Is This IPO Halal?

An interrogation of the Dangote Refinery IPO offer

Alhamdullilah, we are pleased to have found that the Dangote Refinery IPO shares are currently Shariah-compliant - Professor AbdulRazaq AbdulMajeed Alaro, mni, FIFP, FCPA, M.IoD, MCIArb, CIAE, CSA

Nigeria has lately been overcome by what I can only describe as Dangote-mania. The upcoming public listing of Dangote Petroleum Refinery & Petrochemicals (DPRP, but I prefer to use “Dangote Refinery” as I will do throughout this piece) has been described as “The People’s IPO” and yet another example of Aliko Dangote selflessly doing something for Nigerians and Africans across the continent.

As with all things Dangote, his psychological hold on Nigerians (I often say that culturally, Nigerians view the richest man in the country as a kind of god-ordained office) is such that various institutions that you might expect to at least be neutral have openly taken positions in favour of this IPO. As an example, Nigeria’s pensions regulator, as far back as May, issued a directive waiving the applicable operating-history and profitability requirements it would normally demand of pension funds planning to make such an investment. It did this by invoking the refinery’s strategic importance to Nigeria and its “fundamentals” (no numbers had been published at the time.)

Research, Re-search

And then there are the investment banks who have been publishing their “research notes”. The quality of what they have put out in the name of this IPO has truly shocked me. I have managed to go through three such documents and want to highlight the most egregious findings in what Renaissance Capital (operating cost comparison), Chapel Hill Denham (peer average) and Cardinal Stone (dividend yield) have put out. It’s important to point out that all of these firms have commercial relationships with the Dangote Refinery. The IPO prospectus (more on this in detail) names all three of them as joint issuing houses.

The most striking (and hilarious) example of this comes from a disclosure on page 60 of Cardinal Stone’s document. The statement cannot be improved so I quote it in full here: “The content of this research report has been communicated with the Company, following which senior management or relevant executives at the company approved the report. However, no influence was exerted by the management team on the content or recommendations.”

The ‘Company’ here is of course Dangote Refinery and so Cardinal Stone are telling you that what you’re reading was approved for publication by Dangote Refinery, the subject of the research. They also disclose on the same page that the analyst(s) who wrote the report “holds personal positions (directly or indirectly) in a class of the common equity securities of the Company” and that “the analyst(s) responsible for this report, is a board member, officer, or director of the Company.” RenCap are also unusually explicit in their disclaimer. In a report they themselves titled “The People’s IPO”, they said this on page 30: “The Communication is not an advertisement of securities nor independent investment research and has not been prepared in accordance with legal requirements designed to promote the independence of investment research”. Wonders.

Chart 01 above shows how each of the issuing houses are valuing the refinery. The refinery’s own prospectus values it at $46 billion (before issuing new shares) but all the issuing houses are very confident that it is worth a lot more than that. Chapel Hill Denham goes the furthest by saying it thinks the refinery will be worth $113.4 billion by 2030 (perhaps it would have discovered how to refine water into wine by then). Amusingly, CardinalStone’s own comparable-company methods actually put the refinery substantially below the IPO valuation at $27 billion (EV/EBITDA method) but they more than doubled their headline valuation target, because of reasons.

To more substantive matters. On pages 4 and 46 of the Cardinal Stone report, the bank advertises a 39.6% “expected total return” broken down into 31.1% capital appreciation and an 8.5% dividend yield. Now when you go to page 41 of the same document you read this: “On these assumptions, we forecast a PAT margin of 12.9% in FY’26 (H1’26: 13.1%) and profit after tax of $3.8 billion, equivalent to EPS of $0.03 (N44.743)” Two pages down on page 43 it says: “In line with management guidance, we forecast a mean payout ratio of 30.0% between FY’26 and FY’28 as the expansion program progresses”. The reference purchase price is of course N525 per share.

Now taking all the numbers published by Cardinal Stone themselves in their report and doing some simple arithmetic with them, you get this:

This is an unequivocal error which overstates the cash yield by roughly six percentage points and more than triples the dividend yield supported by their own model.Their advertised dividend yield of 8.5% can only be possible if every single naira of earnings is distributed, even though the report forecasts distributing only 30%. All of these contradictions are contained in the report they themselves published.

Moving on to Chapel Hill Denham. On page 31 of their report, they published an emerging-market comparison with 15 available EV/EBITDA observations. Have a look at the column I highlighted below:

When you add up those numbers you get 96. Taking the mean (average, unless it now means something else) gives you 6.4. Somehow they have arrived at 9.5 as their own mean. The surrounding P/E and forward EV/EBITDA averages are calculated correctly so this is clearly an error. The problem with this error is that when you go back a couple of pages to page 29 they say: “On an adjusted peer average of 9.5x and 6.1x EV/EBITDA, we derive an average fair equity value of US$42.89bn…” meaning they are actually using that number in their calculations of the company’s valuation (they don’t explain anywhere what they mean by “adjusted” or anything they adjusted to take the 6.4x to 9.5x.)

Correcting that error alone (while keeping their two-year averaging and valuation weights) reduces its overall equity valuation by approximately $2.21 billion, from $62.53 billion to $60.32 billion. But what is $2 billion among friends?

Finally RenCap. On page 6 of their report, they show this very eyebrow raising chart:

Just before the chart on the same page they say this: “The cost discipline is evident in DPRP’s direct production cost structure translates into a clear competitive advantage when benchmarked against some US independent refiners. Figure 3 compares DPRP’s targeted controllable cash operating cost against the most recent disclosed figures for Marathon Petroleum, CVR Energy, Valero, HF Sinclair, Par Pacific, and Phillips 66, all measured on a consistent basis excluding depreciation and amortisation.”

I read that and was like woah, Dangote Refinery may have discovered the elixir of life here if their operating costs of $0.4 per barrel in 2025 are a tiny fraction of American refiners’. How did they arrive at this figure I wondered. RenCap’s report was published before the IPO prospectus was made available to the public but they clearly had an advanced read of it as one of the joint issuing houses. On page 2 of their report they have a throughput figure of 432,000 barrels per day for 2025. This implies 157.68 million barrels for the year (432,000 x 365). So what operating costs will you give $0.4 per barrel? You are looking for something like $65.5 million ($65.5 million divided by 157.68 million rounds to $0.4). Hmmm, where can we find this number I wonder? Oh look! On page 86 of the IPO prospectus I saw this:

Of all the operating expenses incurred by Dangote Refinery, RenCap decided that the only one that counts in their magic operating expenses number is admin expenses. To give you a sense of how ridiculous this is, I pulled out some operating costs from the IPO prospectus (page 83) in the table below:

If you apply that $1.35 billion to the 157.68 million barrels of throughput for the year you get $8.58 per barrel before adding admin expenses. This will put it above every US refiner on the RenCap chart. And this makes sense given the RFCC issues the company battled with all through 2025

RFCC and Refining: An Explainer

---

These are by no means the only errors or ridiculousness in these reports. For example on page 1, RenCap says: “Phase 2 expansion will double capacity to 1,400kbpd and lifts polypropylene to 2.4mmtpa at a cost of USD 12.4bn which will be financed by the IPO proceeds”. How does an IPO that is raising $1.55 billion finance an expansion costing 8x more? On page 9 of Chapel Hill’s report where it discusses governance aligned with global institutional norms, it highlights Dangote Oil Refining Company’s approximately 66% holding, then notes that special resolutions require 75%. This gives the impression that DORC alone cannot carry special resolutions. A few pages later on page 12, it produced this chart:

Just DORC and Dangote Industries Limited alone add up to 80.73% ownership. What are we doing here? And you’re not going to believe who owns Greenview either.

Cardinal Stone use 112.981 billion shares throughout its valuation and forecast per-share figures. The footnote on page 48 of their report explicitly says these are the shares issued at June 2026. The problem is that in the prospectus, it says that the private placement increased issued shares to 120.129 billion. When you add the 4.1 billion new shares to come from the IPO, you get 124.229 billion shares, which is 10% larger than what Cardinal Stone used. Just changing that denominator alone reduces their total return target from ₦688.09 to ₦625.79. Their model already takes account of the new shareholder cash coming in from the IPO, so you cannot possibly include the benefit of such cash coming in and then leave out the very shares bringing them in from your denominator. Basic.

---

This piece is not really about the investment banks talking their books. I only included the above critique as a way of showing how everyone rallies around Aliko Dangote once he sets about something to the point of sloppiness. No questions will be asked and it can quickly become a suffocating atmosphere for anyone who expresses even mild skepticism. It played out exactly this way in 2013 when the entire banking sector in Nigeria mobilised to help fund the refinery’s construction.

Now let us get into the meat of this post which is the refinery IPO as presented in the 194 page prospectus. There are some useful things to welcome about the IPO. It gives Nigerians an opportunity to own part of an operating industrial business, with the same ordinary share rights as existing investors. The listing also brings the company’s finances and its dealings with other Dangote businesses into greater public view, allowing shareholders and others to examine how it is being run (I have to point out that page 31 incorporates full audited financial statements and their notes by reference but those notes are not all included in the prospectus.) A minimum subscription of ₦5,250 for ten shares makes participation possible for people who would otherwise have little opportunity to invest directly in an enterprise of this scale. These benefits deserve some recognition.

The refinery’s contribution to Nigeria and the merits of buying its shares at ₦525 remain entirely separate questions, however, and each question requires its own evidence.

The ownership calculations assume full subscription to the base offer and no other changes. The 3.30% is the new issue’s share of capital, not necessarily the total tradable free float.

Questions to answer

In this section, I want to try to answer 3 main questions:

1. What are the sustainable earnings that can justify the price at which shares of the Dangote Refnery are being offered?

2. How much of this valuation is based on the existing working refinery (650,000bpd) and how much depends on the planned expansion (1.4 million barrels per day)?

3. There has already been a private placement and refinancing. When you take all those into account, how does it change the picture of what investors are buying with this IPO?

The first big challenge is that the financial presentations need reconciling because different sections report materially different operating cash flow and capital expenditure figures. Secondly, the sustainability of the H1 2026 profit turnaround needs testing because even though the operating improvements are real, favourable product prices also contributed significantly on account of the Iran war. Thirdly, the expansion, debt servicing and dividends are all going to be competing for the same cash so they need to be assessed together.

Starting with the first, which figures should investors use? The reporting accountant’s extracts and the historical financial summaries provide different answers to questions about the refinery’s finances. For the first half of 2026, the extract on page 74 of the IPO prospectus reports $1.513 billion of cash generated from operations. The summary on page 78 reports $1.273 billion. Cash purchases of property, plant and equipment fall from $162.2 million in the first presentation to $33.4 million in the second. Both tables cover the same period and both amounts are expressed in dollars so it cannot be down to exchange rate differences.

One company, many profits

The profit-and-loss account have the same issues. H1 gross profit is $2.583 billion in the reporting accountant’s extract (page 70) and $2.495 billion in the summary (page 76.) Finance income changes from $498.5 million to $49.6 million (perhaps this is a $400m typo?), while finance costs change from $807.6 million to $308.8 million. These differences are offset elsewhere so that profit before tax agrees across both sources ($2.106 billion), but they leave the reader with different pictures of operating profitability and financing costs. The prospectus talks about offsetting “derivative contracts” with Dangote Industries (Note 4.1, page 88), which may explain part of the presentation but there is no schedule connecting these figures.

You can see that the operating profit in the historical summary is $49.905 million lower¹, while its net finance costs are lower by exactly the same amount, allowing both set of accounts to arrive at the same $2.106 billion profit before tax. But the prospectus does not provide a complete explanation of why the same business, over the same six months, has these different allocations of income and expense.

Is this the new normal?

Be all that as it may, $2.1 billion of profits is a very large number and it is what much of this IPO is being sold on. Everyone knows that the Iran war has had a big impact on the fortunes of the Dangote Refinery so I would not belabour that point. The question for me - and I imagine any investor - is: does the prospectus provide any indication or evidence that the earnings level it presents can become normal for the completed refinery?

Stable and full-capacity operation of the refinery only began in March 2026. The operational improvement and the exceptional market conditions caused by the Iran war all arrived together, so the available results cannot cleanly separate the profit attributable to each. Begin with chart 06 below comparing H1 2025 to H1 2026:

The key thing to note from these numbers is that 98.4% of the improvement in pretax profit is accounted for by the increase in gross profit. Everything below gross profit contributed a net improvement of just $38 million. What I’m trying to get to is what exactly explains the move from a loss in H1 2025 to profit in H1 2026. Gross profit, by definition, is what remains from your sales after deducting production costs, including (for a refinery) crude oil, direct labour and depreciation of production assets.

Looking at it another way using the prospectus’ historical summary figures:

In H1 2025, the refinery generated just $145 million of gross profit, against a combined net burden of $427 million from everything below that line. That left it with a pretax loss of $282 million. A year later, the burden between gross profit and profit before tax had fallen by just 9%, to $389 million, but gross profit had risen to almost $2.5 billion. There was now enough gross profit to absorb those remaining costs and leave $2.1 billion before tax. That is where my earlier 98.4% comes from: the $2.350 billion increase in gross profit divided by the $2.388 billion improvement in pretax profit.

We are thus left with a very big question unanswered. The refinery sold a lot more product: petrol volumes almost doubled, while diesel volumes increased 62%. Selling prices also increased quite sharply - approximately 35% for petrol and 78% for diesel. The prospectus says the refinery was better able to absorb fixed costs this year but it does not provide any reconciliation that separates volume and operating efficiency from changes in product prices relative to cost of crude. In other words, while the figures demonstrate improvement, they do not tell us how much of these should be carried into a normal year (with no war.)

A simple way to think of this is that what happens above gross profit is largely out of the control of a refinery (its inputs and outputs are priced in a competitive and transparent international market) while what happens below it is largely within its control.

Speaking of the Iran war (I know I promised not to belabour the point, bear with me), how should we think about this in relation to the Dangote Refinery? A refinery makes money from the difference between what its products sell for and what its inputs cost. In that sense, a fall in crude oil prices can actually be good for a refinery’s profitability if petrol, diesel and jet fuel prices remain high. The IEA said something about this in early July when crude prices fell sharply as supplies recovered, but tight product markets still pushed refining margins to four-year highs. The reverse is, of course, also true.

Let’s quickly go back to RenCap’s analysis (I know I already mocked them but no hard feelings) to see why this is important:

Just a $5 reduction in the margin earned on each barrel removes more than $1.1 billion of annual EBITDA in their sensitivity analysis. What this means is that the refinery can continue to operate successfully while its earnings fall substantially. A very revealing detail is what the various optimistic forecasts by the issuing houses require. RenCap estimates that the refinery’s gross refining margin fell from $33.7 per barrel in Q1 2026 to approximately $18 in Q2 2026. But its full-year forecast somehow requires a recovery to approximately $29 - 30 in H2 2026, followed by $30.20 in 2027. The report explicitly identifies that rebound as one if its key assumptions. Chapel Hill Denham on their part assume $25 in 2027. These are materially different judgements about the earnings capacity of the same asset given what just a $5 swing can do. (Renaissance Capital, p.12; Chapel Hill Denham, p.8.)

Dear reader, I leave you with this question on this point - how much of that additional $2.35 billion in gross profit can the refinery earn again under more ordinary market conditions and what evidence supports the answer embedded in the IPO price? Whatever the answer is, it is not in the prospectus. It only says on page 80 that: “the Issuer estimates that it will be able to achieve a GRM [gross refining margin] of approximately US$24.2 per barrel in 2026” but immediately qualifies this by saying actual margins may vary with crude oil prices and market conditions.

Cash and Debt

When you look at H1 2026 numbers, you are presented with a very reassuring number in terms of how much debt the refinery is carrying.

But the key thing to note here is that this low-debt picture is a snapshot taken after the company received substantial fresh equity and before expansion spending ramps up. This is important because it has a big impact on how much cash shareholders can receive while the company is spending on doubling the refinery’s capacity to 1.4 million bpd.

The numbers in the prospectus are a bit confusing so I’m hoping chart 10 below will make it a bit clearer:

The summary of this table is that the refinery received $2.71 billion from share issues and deposits for shares before the IPO. This is equivalent to about 85% of the increase in its cash during H1 2026. From the prospectus, it is not clear which share transactions account for $468.7 million out of those receipts, but the cash-flow statement recorded that the money came in. This fundraising helped to boost the refinery’s cash balance to $4.27 billion by June. The refinery still owed $5.67 billion, but subtracting that cash is how they got to net debt of just $1.40 billion in chart 09. That low net-debt figure is therefore mostly down to cash supplied by investors. As that cash is spent on the planned expansion, it will of course no longer be available to reduce the net-debt figure unless operating cash flows or further equity funding replenish it.

This is, of course, a perfectly legitimate way for a company to strengthen its balance sheet. The question here is what happens when the money has been spent. Simply paying for construction cost will reduce cash and increase net debt as a result, even without taking out another loan. That is to say, that cash deducted when calculating net debt is also part of the money available to fund expansion. It cannot serve both purposes - debt reduction and construction - indefinitely.

What makes this even more important is the spending timetable disclosed in the prospectus. $4.8 billion of capital expenditure is budgeted for the remainder of 2026 (i.e. H2 2026), followed by $3.9 billion in 2027 and $3.1 billion in 2028. That adds up to $11.8 billion over the next two-and-a-half years. The IPO’s net proceeds translate to approximately $1.55 billion at the exchange rate used in the offer, which works out at about a third of the planned H2 2026 expenditure alone. (Prospectus, pp.34, 93 and 156.)

I don’t want to keep punching at the investment banks and their research reports (what am I saying, of course I want to) but this capital expenditure point is another part of their research that makes no sense.

By what logic have these guys all come up with a full year’s capital expenditure spending that is less than half of what the refinery itself says it will spend in just six months of the year? Someone who is good at spending please explain this to me in the comments.

Now to be perfectly clear, I’m not suggesting that the company is going to run out of cash or anything of the sort. The company does have other resources available to it. Alongside that $4.3 billion of cash it disclosed in June, a $750 million bond issue was completed in July alongside the $258 million private-placement tranche in chart 10. Adding those amounts to the expected net IPO proceeds of $1.55 billion brings us to approximately $6.82 billion (before bond issuance costs and other cash movements) after June. This is more than enough to cover the planned H2 capital expenditure, leaving about $2.02 billion before subsequent operating cash flows, debt service and other uses. The key point is that a big chunk of that cash has a planned use and by definition will not be available to pay shareholders. Am I communicating?

Control

I want to talk about something that has increasingly become normalised in Nigeria around the way “public” companies are owned and governed. Just because this absurdity has been going on for a long time (Dangote himself “pioneered” it with his cement listing in 2010) does not mean everyone should shrug and accept it as the mandate of heaven.

A fully subscribed base offer would leave Aliko Dangote with approximately 84.39%² ownership of the refinery, assuming his existing shareholding remains unchanged. The new IPO shares would represent 3.30% of the enlarged company, with existing minority shareholders holding the balance. This large holding gives Dangote a financial interest in the refinery’s success but the public will actually be buying shares in one company that sits inside a larger group. If, for example, another Dangote company supplies gas or cement to the refinery, the price determines how much profit each company earns. A refinery’s shareholder will not have any claim on the supplier’s profit just because both businesses carry the Dangote name. These are among the supplies the prospectus lists under transactions with related parties, therefore the terms on which all the Dangote companies deal with each other has an impact on the value of the shares being offered (Prospectus p.158).

The financial relationships between the Dangote companies are much deeper than just buying supplies. The refinery repaid a $3.99 billion loan to Dangote Industries Limited (DIL) during H1 2026. DIL is also the counterparty to the arrangement under which losses on the refinery’s external commodity derivatives were offset by a gain on an intercompany derivative (if this sounds complicated, that is the point.) There was no net derivative loss for H1 2026 shown in the prospectus because of that offset. So in this particular case, we can see that the group arrangement helped to support the refinery’s reported profit. I have no idea on what the pricing and duration of these derivatives are, or the conditions under which it can be changed, and neither do you. But you have to factor it into any discussion of the sustainability of the refinery’s profits (pp.78, 88, 92–94 and 159).

The refinery’s Finance and Investment Committee has the responsibility for reviewing these financing strategy decisions and overseeing treasury activities. The three members of the committee are:

Every member of the Finance and Investment Committee also has a role at DIL. The director described as “independent” at the refinery has a seat on DIL’s board while DIL’s vice-president chairs the refinery committee responsible for overseeing financial reporting and internal controls. These overlapping responsibilities make the arrangements for scrutinising any transactions the refinery has with DIL particularly important. What exactly does “independence” mean when the company on the other side of a transaction is one that that the committee members also hold senior roles in?

There is also a big difference between establishing committees and show how they work in practice. The prospectus says that the newly constituted board committees had not held any meetings as at the date the document was published. A separate Statutory Audit Committee is expected to be constituted after the listing, with shareholder representatives elected at the first annual general meeting. It also says governance policies are “being reviewed” and that any such policies still awaiting formal adoption will be submitted for board approval (without identifying which policies remain outstanding.) Investors are therefore being asked to assess safeguards whose operation is still to be demonstrated (Prospectus pp.130–132).

As I say, this is all very “normal” in Nigeria where owners of “listed” companies own almost all of its shares and control it via various opaque structures. But just because something has been normalised in Nigeria does not mean it should be accepted with a shrug.

Peer Review

We come to the final point I want to make about the prospectus and the refinery’s valuation. A standard practice is to compare a company like the Dangote Refinery to similar (in terms of operations, size etc) refineries from around the world to get a sense of what its value ought to be. None of this is an exact science but you can help yourself a great deal by the choice of peers you select.

The way that the three investment banks have gone about this part of their exercise has been quite problematic to say the least. RenCap did a 19-company peer group which included Japan’s SALA: a company whose businesses span gas supply and car sales, but no disclosed crude-oil refining operation. Removing SALA doesn’t really change the valuation benchmark but it makes me wonder who checked the list. There is an even bigger problem with HD Hyundai, whose refinery subsidiary includes shipbuilding and other businesses, and SK Innovation, which combines refining with batteries and other energy operations. These are a different collection of businesses from the one being offered by Dangote Refinery through this IPO. Valero, Marathon Petroleum and Phillips 66 account for almost 70% of the enterprise value in RenCap’s table. When you add the enterprise values and earnings you get a multiple of 7.74x compared with a median of 6.27x across the individual companies. It is not that this approach is wrong per se but the weighting affects the answer you arrive at and needs a justification (RenCap, p.10).

CardinalStone’s is better and names such as Tüpraş and Valero are worth retaining. But its unadjusted comparisons also include Marathon and Phillips 66 (for the reasons already stated above), and Bangchak, which has upstream and power businesses. More importantly, its table does not identify the earnings periods behind the peer multiples and yet it applies them to Dangote’s next twelve months of forecast earnings. The median is arithmetically fine, and its relatively low peer valuations are not the main reason for its highly optimistic conclusion, which comes from the much higher discounted-cash-flow valuation which carries 70% of the weight (CardinalStone, p.49).

Chapel Hill Denham has the more elementary obstacle that we already discussed in the opening section of this piece. It supplies “adjusted” peer averages of 9.5 times EBITDA for 2026 and 6.1 times for 2027 but it does not even name the companies in this peer group let alone explain the weights and adjustments. There may be a reasonable explanation behind those numbers but the research note does not give readers any information needed to test it (Chapel Hill Denham, p.29).

To get around these problems, I have come up with a shorter and, I hope, more useful peer group of four companies (you’re welcome.) Three are worth retaining from RenCap’s and CardinalStone’s lists while Mangalore is my own addition. I have included each one for a specific reason because the comparison should help us assess the specific business being offered to investors - in this case the Dangote Refinery - including the expansion it still has to finance. There is no need to include companies on the list just because they have “oil” in their name.

Mangalore Refinery and Petrochemicals (MRPL), India operates a single coastal refinery with a product mix very similar to Dangote’s. Petrol, diesel and jet fuel accounted for approximately 84% of its turnover in 2025–26 with polypropylene contributing another 4.4% (PDF, pp 109-110.) That is a pretty good match for Dangote Refinery whose current earnings depends almost entirely on producing transport fuels. MPRL’s less obvious attraction is its ownership structure. ONGC (Oil and Natural Gas Corporation Limited, India’s state-controlled oil and gas producer) and HPCL (Hindustan Petroleum Corporation Limited, a major Indian refiner and fuel retailer) together own 88.58% of MPRPL, leaving public investors with a small minority interest (are Nigerians and Indians blood relations?) MRPL therefore lets us compare Dangote Refinery to what investors pay for a refinery whose controlling shareholders remain firmly in charge. It’s also a cleaner comparison because, just like buying Dangote Refinery’s shares don’t entitle you to participate in DIL’s profits, buying MRPL shares only gives investors a claim on MRPL’s business, without participating in ONGC’s upstream oil operations. MRPL is smaller than Dangote Refinery and the state control aspect brings different policy risks, but those are differences we can identify and discuss.

S-Oil, South Korea has refining capacity of 669,000 barrels a day at its Onsan complex, close to Dangote’s rerated 650,000 (or the rerated 700,000.) This is the strongest physical comparison in the group which is again a large coastal complex combining refining with related chemical production. It gives us the concentration of a substantial business in one location. An important difference to acknowledge is that lubricants contributed approximately 29% of S-Oil’s H1 2026 operating profit (PDF, page 14), so its valuation is a bit more than just a pure fuel-refining multiple. Its Shaheen petrochemical expansion nevertheless makes it very useful. Commercial operations at the expansion are targeted to start in early 2027 (PDF, page 12), so, just like Dangote’s investors, S-Oil’s investors also have to value tomorrow’s production while today’s business finances the construction.

Tüpraş, Turkey is the most useful comparison for the pricing argument I will discuss in the concluding part of this piece. Its financial statements explain that selling prices follow Mediterranean petroleum-product prices and the dollar exchange rate (PDF, page 43). Here is an important domestic refiner whose local customers remain connected to international fuel markets. That gives us a very concrete (pun intended, I’m afraid) way to examine the value of a strong domestic position under international pricing and not just saying “emerging market”. Tüpraş operates four refineries, and its broader investments and Turkey’s hyper-inflation environment require us to be very careful when interpreting its multiple. But the commercial mechanism is directly relevant to Dangote and Nigeria.

Valero, United States provides an established benchmark in the international fuel markets into which Dangote sells some of its products. Its Gulf Coast and North Atlantic operations make the connection more useful than just choosing any large American refiner. Its business is also more focused on transport fuels than a diversified energy conglomerate (although renewable diesel and ethanol also contribute meaningfully). Another useful point is that it makes us consider how valuable geographical diversification really is. Dangote’s concentration in one enormous complex can produce economies of scale, but a network offers protection against an interruption at any one location (see earlier article on RFCC I linked above). Valero owns 14 petroleum refineries across the US, Canada and the UK. Low costs are very attractive but so is having another refinery working when another one breaks down.

Having done all that explaining, here is my own peer-group comparison:

The EV/annualised H1 EBITDA column shows how much each business in the peer-group costs (after adjusting for debt and cash) for every dollar of earnings before interest, tax, depreciation and amortisation. Dangote costs $8.94 against a peer median of $5.62, a 59% premium on its peers. The Equity/annualised H1 profit column looks at what shareholders pay to access the remaining profits after those charges. Here, Dangote’s 13.13 times compares with a peer median of 9.13 times, a 44% premium. Dangote is the most expensive company to buy in this group on the first measure but Mangalore is more expensive on the second. The result is that investors in this IPO are being asked to pay a lot more for each dollar of Dangote Refinery’s earnings.

To make these comparisons I have doubled their H1 2026 earnings which might be different from what a full year might look like (there might be some seasonal variation for instance.) If for example I removed S-Oil’s inventory gains and Mangalore’s exceptional repricing income, the operating-earnings premium would reduce from 59% to about 34%.

But the real implication of the numbers in chart 13 is this - If Dangote repeated its first-half performance for the rest of 2026, its annual earnings would be about $5.3 billion. At the valuation implied in this IPO, those annual earnings would need to rise to about $8.4 billion for investors to be paying the same amount for each dollar of earnings as they do for the typical company in this peer group. That means roughly 60% more earnings just to bring its valuation in line with its peers.

This is where the planned expansion to 1.4 million barrels a day enters the discussion. It is the reason you’re being asked to pay that price today and the premium suggests a lot of future success (the construction will go to plan, the RFCC issue will not repeat itself and so on) is already baked in. The prospectus puts the expansion programme’s cost at approximately $14.3 billion (prospectus page 34). That funding must come from the business’s cash generation or additional financing (since the IPO is only going to cover a small part of it as discussed). Again, cash committed to construction is unavailable for dividends and borrowed money carries its own financing bill. All of this is to say that the promised expanded refinery will have to arrive well before any return for shareholders. How much of that expansion represents extra upside for the buyer at ₦525, and how much is the business growing into the price already being charged?

I don’t know, you tell me.

Aliko Dangote (and cement)

I can’t close this piece without talking about Aliko Dangote himself as the promoter and larger than life person behind this IPO. I don’t think any other Nigerian could cause so many to lose themselves in excitement in the way this IPO has done (was it really necessary for RenCap to say on page 20 of its report that “DPRP processes more crude than the combined operational output of every other refinery on the African continent” when Algeria and Egypt alone processed approximately 1.17m barrels per day in 2024?)

As a very self-satisfied Dangote Detractor I will not be investing in this IPO. Anyone who knows me knows that development is my favourite topic. To me, Dangote is anti-development and as such I’m opposed to his businesses in principle and in practice. If this IPO is the one sure route for me to become rich (something I’d very much like to happen), then I’m going to have to remain poor. But you can have pity on me by taking out a paid subscription to 1914 Reader to help mitigate this risk of poverty.

Cement, as I will never get tired of saying, is a key to civilisation. There is no path to development that does not run through pouring lots and lots of concrete. Forget even housing, you need it to build drainage systems, power plants, roads, bridges and to overcome the harsh external environment in a way that makes it habitable for humans to flourish in it. It has taken me living outside of Nigeria for two decades to really appreciate just how difficult the physical Nigerian environment is. Cement is one of the biggest keys to getting out of that.

A World Bank study on cement usage in Mexico found this:

In this paper, we help to fill this gap by investigating the impact of a large scale effort by the Mexican Government to replace dirt floors with cement floors on child health and adult happiness. We find that replacing dirt floors with cement floors significantly improves the health of young children. Specifically, we find that a complete substitution of dirt floors by cement floors in a house leads to a 78 percent reduction in parasitic infestations, 49 percent reduction in diarrhea, 81 percent reduction in anemia and a 36 to 96 percent improvement in cognitive development. Additionally, we find that replacing dirt floors by cement floors significantly improves adult welfare, as measured by increased satisfaction with their housing and quality of life, as well as significantly lower rates of depression and perceived stress.

If Aliko Dangote’s stewardship of this critical civilisational tool has produced anything, it is unhappiness. His singular obsession has been to charge as much as possible for the product without any thought about its effect on Nigeria’s development, even though it is the country that has made his fortune possible. No amount of support he is given changes this - he will take it as his god-given right to bigger margins. And we can illustrate this with coal.

A cement plant uses a lot of energy to generate heat. It burns fuels such as coal or gas to heat limestone and other raw materials to around 1,450°C in a kiln, which turns them into clinker, which is then ground into cement. It is pretty much its biggest cost input. If I go back as far as the Dangote Cement 2013 results (PDF, page 5), I can see that Obajana - Africa’s largest cement plant by installed production capacity (it can turn out 325 million bags of 50kg cement a year) - used practically no coal at all at that point. Fast forward to 2020 (this is the latest disclosure for Obajana specifically I can find) and coal had become 50% of its fuel mix (PDF, page 50). The reason for this shift is that around 2015, after Dangote suffered some disruption to its gas supplies and cost of importing fuel increased (something other businesses in Nigeria also suffered), the Nigerian government stepped in to award him licences to a couple of very large coal mines with decent quality coal in them (PDF, page 29). I have not found any evidence that Dangote paid anything meaningful for these coal mines. For a sense of scale, those two mines alone (there have been others - link opens as an Excel spreadsheet) cover almost 90% of the size of Paris city.

The point I made earlier about dealings between Dangote entities is useful here. The arrangement is that a company called Dangote Coal Limited mines the coal and sells it to Dangote Cement at whatever price they determine between themselves. We have no idea what it costs Dangote Coal to produce the coal for sale since that company’s accounts are a black hole not visible to the public. But we can get an idea. On an earnings call last year, the CFO of Dangote Cement said: “We have started using local coal. Three, four years ago, we were importing expensive coal at about $80-$100 per tonne. As at today we are using local coal and buying for less than N30,000 N40,000 per tonne.” In other words, one Dangote company sells coal to another Dangote company for roughly $20 - 26 per tonne (using 2025 exchange rates). As of today, on the international market, coal sells for just under $150 per tonne. I stress again, this is the price that one Dangote company sells the coal to another Dangote company. Whatever the profit on the sale ends up with Dangote anyway. And the price is still a fraction of what it costs on the international market.

What have Nigerians benefitted from making it easier for Dangote to produce cement in Nigeria? As I have previously demonstrated, however you slice or dice it, Nigeria’s per capita cement consumption has gone down from where it was in 2013 to today, even though Nigerians barely consume half of the cement that Indonesians do. As for Aliko Dangote himself, in 2008 (the year he made his debut on the list), Forbes reported his net worth at $3.3 billion. As of today, the Forbes tracker lists his net worth at $31.2 billion. I’m here to tell you that it is possible for someone to get richer as people buy fewer of his products. To misquote the Bible - “He must increase, but I must decrease.” All of the government support, all of the tax breaks, all of the import bans - how has Aliko Dangote expressed his thanks to the country that has made him so fabulously wealthy? Revenue per tonne of cement in China works out at $34 per tonne (China Resources), in India $66 per tonne (Ultra Tech), in Indonesia $60 per tonne (SIG). And Dangote in Nigeria? $111 per tonne (₦167,186 in naira).

There is no mystery to this other than that Dangote sees the biggest possible margins obtainable from Nigeria and Nigerians as his singular obsession and god-given right. This is useful to bear in mind for the upcoming IPO for the reason I alluded to above in the discussion about Tüpraş. In the IPO prospectus, I counted four mentions of import parity:

It is very important for everyone to be clear about what this means given that Dangote has been very clear about it and given his antecedents. Nigeria may be reducing its dependence on imported fuel with this refinery but it will continue to import fuel-price shocks. Import-parity pricing uses the cost of obtaining an equivalent imported product as a benchmark. In general, that means the international price of fuel plus the costs of bringing it into the Nigerian market. It does not matter whether the local crude and delivery do not incur the same international shipping and insurance costs - that is the price he will charge Nigerians for it. No amount of support with naira for crude or operating out of a tax free zone will change this (on page 113 you can even see that the prospectus places import-parity pricing alongside the advantages of local logistics. It identifies efficiencies from producing locally, but does not specify how much of those savings will reach customers through lower prices. I can tell you for free that the answer is zero.)

Dangote has been clear on this, you - as a potential investor - also need to be clear. If some crazy guy in the Middle East flies a drone into a refinery in Saudi Arabia and oil prices spike, you as a Nigerian will feel that price shock. Dangote will give it to you, whether you want it or not. Just as with cement where he has priced the product at what you might get it for in China plus the cost of shipping it to Nigeria (which is very large due to the weight of cement), this limits any consumer-price protection Nigerians can expect from having a local refinery on their doorstep.

As I said before and will say again - I object to Aliko Dangote and his businesses in principle and in practice. If I were to be building something in Nigeria today, I will go out of my way (including paying more) to avoid using Dangote Cement.

---

The day before Tim Cook handed over to John Ternus as the new CEO of Apple, the Financial Times did a piece on his (Cook’s) legacy in numbers. It worked out that Apple’s shares had risen by more than 2,000% under his leadership and the company had returned well over $1 trillion in dividends and buybacks to shareholders. All of this happened as the company sold 3.1 billion iPhones under his guidance and went from 60,000 to 166,000 direct employees. Tim cooked and everyone ate.

Over the period he was CEO, he was a paid a total of $880 million in salary, stock awards and other cash incentives. As of today, he owns just 0.02% of Apple. For every $1,000 he added to Apple’s market value, he took home 20 cents for himself. Against the cash Apple returned to shareholders through dividends and buybacks, his compensation was less than 90 cents for every $1,000 distributed.

Dangote is the direct inverse of this. Any value he creates is captured at source by him. No technology boon will accrue to Nigeria from him. No one will leave any of his companies to set up a business elsewhere in the economy that boosts development (you will notice that the cement company only ever makes noise about rewarding “distributors”). Bear all of this in mind as you invest in the IPO (if you make money, you can share some with me out of the goodness of your heart.)

I can only properly express all of this in pidgin English to close: Dangote no dey chop remain for anybody.

1. The operating review identifies the $37.128 million other gains as foreign-exchange gains (pp.87–88). “Not separately shown” does not mean those gains were absent from the underlying accounts. As mentioned, the prospectus discusses offsetting derivative contracts, but does not provide a numerical bridge between these presentations.

2. Calculated from Dangote’s disclosed beneficial holding of 104,834,654,430 shares, divided by 124,228,915,901 shares after issuing the 4.1 billion IPO shares. I’m assuming no additional shares are allotted to him. The 3.30% figure refers to the new IPO shares. Sources: IPO prospectus, pp.33 and 157.