A barrel is processed twice at Tüpraş. First, it passes through the refinery lines in İzmit, İzmir, Kırıkkale and Batman. In the first three months of 2026, the company charged 7.090 million tonnes of crude oil and semi-finished products and produced 6.776 million tonnes of saleable products. The second passage is through the contract book. As of 31 March, there were forward crude oil purchase and sale transactions covering 80.925 million barrels, along with 5.445 million barrels of crack-margin hedges for the second quarter.
The second barrel is not a gambling chip. The financial statements say the crude oil transactions are used to hedge the price risk of future sales, the crack transactions to protect the product margin, and that there was no ineffective portion in hedge accounting. Tupras Trading in London also manages the process of hedging the price risk of inventories. Yet the scope stated in the source for the 38.825 billion TL block is broader: the note links this money to all derivatives transactions carried out on overseas exchanges, without allocating it one-for-one only to hedging contracts.
That distinction deserves the headline. Tüpraş’s balance sheet shows 129.900 billion TL in cash, while period-end cash under the cash-flow statement definition is 76.386 billion TL. Of the difference, 38.825 billion is blocked for derivatives transactions, 14.228 billion for the regulatory revenue share; the remainder is a small electricity-market block and accrued interest. Blocked money is not lost. But money that cannot move and money that can be freely distributed are not the same thing.
The question is not whether the refinery is running. It is. The investment question is this: how much freedom does a 515.9 billion TL market value buy from this treasury, and how many years does it buy from this cycle?
The furnace is fast, white products a little thriftier
The first-quarter operation cannot be dismissed as weak. Capacity utilization rose from 83.2% a year earlier to 94.5%. Production increased 15% to 6.776 million tonnes, while total sales also rose 15% to 7.376 million tonnes. Domestic sales grew 20%. Diesel sales rose 24%, and jet-fuel sales 16%. Ditaş’s tankers, Körfez Ulaştırma’s trains and the trading desk in London are feeding the same physical flow.
| Metric | Q1 2025 | Q1 2026 |
|---|---|---|
| Total charge | 6.240 | 7.090 |
| Production | 5.914 | 6.776 |
| Total sales | 6.395 | 7.376 |
But what is inside the volume matters. White-product yield fell from 84.4% to 82.8%. Fuel-oil production rose 67%, and bitumen production 40%. This is not, by itself, proof of a bad quarter; demand, maintenance and crude quality can change the mix. Still, it is a reminder that high utilization does not cure low margins. Producing more tonnes does not mean that every tonne is as valuable as diesel, jet fuel or gasoline.
The financial results carried the vitality in the volumes. In 31 March 2026 purchasing power, revenue rose from 207.6 billion TL to 258.3 billion TL, and operating profit from 5.1 billion TL to 11.4 billion TL. EBITDA reported by management increased from 12.6 billion TL to 16.727 billion TL. Net income attributable to the parent company leapt from 127 million TL to 3.710 billion TL.
The last number is less comfortable than the first two suggest. Taxes took 6.568 billion TL of the 10.383 billion TL pre-tax profit. Of that, 2.160 billion TL was deferred tax expense. A reduction in the expected future use of tax advantages from investment incentives generated 5.419 billion TL of deferred tax expense. TMS 29 also recorded a 2.462 billion TL net monetary position loss. “Profit rose 29-fold” is true; it is incomplete without saying that the profit carries a heavy tax line and inflation accounting underneath it.
Two locks on the treasury
Treating 129.900 billion TL as one undivided treasury is an expensive mistake. Of this, 74.062 billion is time deposits and 2.324 billion is demand deposits. In addition, 38.825 billion TL is blocked on demand for derivatives transactions on overseas exchanges, 14.228 billion TL is blocked as the national stock revenue share, and 1.7 million TL is tied up for the electricity market. After deducting 459.7 million TL of accrued interest, period-end cash under the cash-flow statement definition is 76.386 billion TL.
That 76.386 billion TL does not carry an economic-freedom label. It is an accounting definition; by itself, it does not prove that every lira can be unconditionally directed to dividends or investment tomorrow. An economic reading takes two steps: blocked amounts should not be counted as distributable treasury, but neither should they be erased as if they were losses. The company’s own net-debt calculation arrives at 21.651 billion TL of net cash, deducting 55.194 billion TL of debt from 76.845 billion TL of cash and financial investments, including accrued interest. That is the number used in valuation; the number that describes liquidity tension is the 38.825 billion TL block.
| Item | TRY bn | Public definition and economic reading |
|---|---|---|
| Balance-sheet cash | 129.900 | Headline amount in the statement of financial position |
| Blocked for all derivatives traded on overseas exchanges | 38.825 | Demand blocked deposit; not allocated one-for-one solely to hedging transactions |
| Blocked regulatory revenue share | 14.228 | National-stock revenue whose use is under the regulator's discretion |
| Other blocked cash and interest accrual | 0.461 | Outside the cash-flow statement definition |
| Period-end cash under the cash-flow statement definition | 76.386 | An accounting definition, not standalone proof of economic freedom |
| Cash and investments in the net-debt calculation | 76.845 | Company definition in Note 29 |
| Financial debt | 55.194 | Debt side of the net-cash bridge |
| Company-defined net cash | 21.651 | Valuation input |
The revenue share is a separate lock. This money, added to fuel prices, is collected to finance the national stock obligation; its use is at the discretion of EPDK, and a liability of roughly 14.316 billion TL sits on the other side of the balance sheet. This is not Tüpraş’s transformation budget.
Across from the derivatives block stands a large book. Against 25.347 billion TL of derivative assets there are 62.127 billion TL of derivative liabilities, producing a net fair-value deficit of 36.780 billion TL. The cash-flow hedge reserve is negative 16.210 billion TL in equity. During the period, the after-tax expense effect of this reserve reached 17.201 billion TL, and the company recorded a total comprehensive loss of 13.880 billion TL despite 3.815 billion TL of net profit.
Adding these three amounts on top of one another is wrong. Blocked deposits, the balance-sheet value of derivatives and the equity reserve are different cash and accounting faces of the same structure. Ignoring them is also wrong. Hedging transactions can soften economic risk while collateral moves liquidity and fair value moves equity.
Cash flow shows the tension in plain sight. Period profit was 3.815 billion TL, while operating cash flow was 6.288 billion TL. At first glance, profit appears confirmed. But from 42.393 billion TL of cash generated from operations, 35.097 billion TL of “other cash outflow,” including the change in blocked deposits, disappeared. After the 4.150 billion TL spent on purchases of tangible and intangible assets, simple first-quarter free cash flow was 2.139 billion TL. This is not normal earning power; it is a quarter snapshot suppressed by collateral movements.
The dividend has gone out, summer maturities are ahead
Tüpraş approved a 33 billion TL dividend from its 2025 profits. The first 20 billion TL was paid in March; the second installment of 13 billion TL is scheduled for the 30 September-2 October payment window. The total distribution equals 6.4% of the 10 July market value. An investor buying the shares today will not receive the first installment; the second installment ahead represents approximately 2.5% of the same market value.
The size of the dividend is not, by itself, proof of quality. The 20 billion TL paid in the first quarter was more than nine times the 2.139 billion TL of simple free cash flow affected by the period’s collateral movements. This is not a payment crisis, but the use of liquidity accumulated in the past. Still, a 33 billion TL distribution and roughly 700 million dollars of annual investment are reaching into the same treasury. The propylene project has a total budget of 271 million dollars, a target of 180 thousand tonnes of annual production, and a commissioning year of 2027. Entek is also borrowing for wind projects with storage and solar projects in Romania.
The debt wall is not frightening today, but it has a name on the calendar. The 4.0 billion TL, 727-day bond issued by Tüpraş on 23 July 2024 matures on 20 July 2026; the 2.8 billion TL, 730-day bond issued on 12 August 2024 matures on 12 August 2026. The total 6.8 billion TL of summer maturities is a small portion of the 76.386 billion TL of period-end cash under the cash-flow statement definition. It is not a threat on its own. The threat is the simultaneous demand for free treasury from dividends, investment, collateral and debt repayment.
The green display has not yet carried the fuel oil
Tüpraş increasingly describes itself as an energy-transition company. Solar projects in Kırıkkale and Batman, Entek’s hydroelectric and wind portfolio, the 20 MW electrolyzer project and preparations for sustainable aviation fuel in İzmir show that this narrative is not empty. Zero-carbon electricity capacity, including the refineries, reached 418.9 MWe.
But the first-quarter 2026 account still smells of oil. Of consolidated revenue of 258.254 billion TL, 256.454 billion came from refining. Electricity-segment revenue fell from 2.949 billion TL to 1.800 billion TL; its operating result turned from a 57.7 million TL profit to a 223.4 million TL loss, and its net result became a 495.5 million TL loss. The technical infrastructure for SAF may be ready, but production depends on “appropriate commercial conditions and demand.” This is a conditional option, not delivered revenue.
OPET is a more tangible second chimney. The carrying value of Tüpraş’s 41.67% stake is 18.000 billion TL. OPET has 1,995 stations and a 20.4% market share in white products. Yet OPET made a 199.8 million TL loss in the first quarter, of which 83.3 million TL fell to Tüpraş’s share. At the same time, OPET distributed a 1.250 billion TL dividend. The distribution network is valuable; its current profit contribution is negative.
This relationship is also a matter of concentration and governance. Tüpraş’s sales of products and services to related parties were 47.244 billion TL, or 18.3% of revenue; 41.830 billion of that was sales to OPET. Koç Holding reduced its direct stake from 6.35% to 4.27%, while control continued through a total 50.67% stake with Enerji Yatırımları. The single state-owned C-group share carries approval rights over decisions that could affect fuel supplies to the Turkish Armed Forces. The minority investor receives neither an automatic alarm nor an unqualified seal of trust: a significant share of sales goes to the group ecosystem, control remains in place, and strategic decisions carry a public-sector lock.
The 104.924 billion TL of guarantees provided looks heavy at first sight. The detail is fairer: 102.917 billion TL is for the company’s own account, and 2.006 billion TL is in favor of consolidated subsidiaries; there are no additional guarantees for the parent, other group companies or third parties. Short- and long-term employee provisions total 3.678 billion TL, while the litigation provision is only 28 million TL. The large risk is not in a courtroom file, but in the possibility that margin and collateral treasury turn against the company at the same time.
A price between six dollars and eight years
At a price of 267.75 TL on 10 July 2026, Tüpraş has a market value of 515.900 billion TL. After subtracting the company’s 21.651 billion TL of net cash, enterprise value is 494.248 billion TL. Multiplying first-quarter EBITDA by four produces 66.908 billion TL and an EV/EBITDA of 7.39x. This is not a one-year forecast. It is a rough control measure that repeats one high-volume quarter four times.
The real bridge should be built from management’s own 2026 sentence: a net refinery margin of 6-7 dollars per barrel, 95-100% capacity, approximately 29 million tonnes of production, 30 million tonnes of sales and approximately 700 million dollars of investment. First-quarter capacity was just below the range, the simple annualized production pace of 27.1 million tonnes was below the target, and the sales pace of 29.5 million tonnes was close to the target. First-quarter investment of 92.7 million dollars was only 13.2% of the annual budget; spending will accelerate.
When I convert 29 million tonnes into barrels using the explicitly stated 7.33 barrels-per-tonne model conversion, and into lira using the 44.77 TL per dollar derived from the first-quarter investment’s dollar and TL equivalents, three earnings cases emerge. A 6-dollar margin produces a 57.1 billion TL refinery EBITDA proxy, 6.5 dollars produces 61.9 billion TL, and 7 dollars produces 66.6 billion TL. I do not multiply the 30-million-tonne sales target by margin a second time; that number is a delivery check, testing whether production can be sold without remaining in inventory and alongside the trading flow. Adding the two volumes would count the same economic barrel twice. This is not a consolidated earnings forecast; it is a working measure linking net refinery margin to volume. Since electricity was loss-making in the first quarter, I am not assigning a positive contribution to the side businesses.
| Case | Margin | Refining EBITDA proxy | USD 700m investment | Normalised cash tax | Firm free-cash-flow proxy | Yield on current enterprise value |
|---|---|---|---|---|---|---|
| Low end of guidance | USD 6.0/bbl | TRY 57.1bn | TRY 31.3bn | TRY 10.1bn | TRY 15.6bn | 3.2% |
| Midpoint of guidance | USD 6.5/bbl | TRY 61.9bn | TRY 31.3bn | TRY 11.3bn | TRY 19.2bn | 3.9% |
| High end of guidance | USD 7.0/bbl | TRY 66.6bn | TRY 31.3bn | TRY 12.5bn | TRY 22.8bn | 4.6% |
This model has two honest flaws. The barrel-to-tonne conversion can vary with product intensity. I am also deducting the entire 700 million dollars from a single year’s cash; a significant portion is growth and transformation investment, not maintenance. The model therefore understates future returns while fully counting today’s cash burden. Even so, before calling the stock “cheap,” this is the cleanest way to see how much cash the price actually buys.
The multiple range should not become a story about historical peers. The 6x, 7x and 8x cases are not presented here as Tüpraş’s historical average. Each shows how many years of gross EBITDA equal enterprise value. If the 700 million dollars of investment absorbs 54.9% of EBITDA at the low margin, a six-year duration gives the cycle and capital intensity their due. Seven years buys one more year of earnings without granting full quality credit. Eight years can be carried only if guidance, the release of collateral and the growth character of the investment can all be defended together.
Most importantly, 8x is no longer a label chosen because it is “optimistic.” At the current price, the market is already paying 7.99x the 61.9 billion TL EBITDA proxy at the midpoint of the 6.5-dollar guidance. Eight years is the market’s demand today.
| Margin case | 6 years of EBITDA | 7 years of EBITDA | 8 years of EBITDA |
|---|---|---|---|
| USD 6.0/bbl | TRY 189.0 | TRY 218.7 | TRY 248.3 |
| USD 6.5/bbl | TRY 203.8 | TRY 235.9 | TRY 268.0 |
| USD 7.0/bbl | TRY 218.7 | TRY 253.2 | TRY 287.8 |
| Current price | TRY 267.75 | TRY 267.75 | TRY 267.75 |
After adding 21.651 billion TL of net cash, a 6.5-dollar margin and eight years of EBITDA produce a per-share value of 268.0 TL, almost exactly the screen price. At the same margin, if duration falls to seven years, value drops to 235.9 TL. At the upper margin and eight years, value is 287.8 TL, producing 7.5% upside to the current price. Even at the low margin, eight years leaves value at 248.3 TL. The market considers hitting the midpoint of guidance insufficient; it also wants a long-duration credit on those earnings.
The second test is the balance sheet. Equity attributable to the parent is 353.671 billion TL; the current price is 1.46x book value and carries a 162.229 billion TL premium over book. That book already includes the 33 billion TL distribution decision and the negative 16.210 billion TL hedge reserve. Deducting them again would count the same burden twice. After subtracting net cash and OPET’s 18.000 billion TL carrying value, the market is assigning 476.248 billion TL of residual equity value to refining, Entek and the other businesses. This is not a fire-sale price.
The strongest counterargument is clear. The derivatives hedge the physical barrel, there is no ineffective portion, and blocked money may be released when its time comes. The Q1 x4 check at 66.9 billion TL is close to the upper-end guidance proxy. If the margin remains above 7 dollars, capacity exceeds 95% and 30 million tonnes of sales are achieved, today’s 8x midpoint calculation will quickly look cheaper. Net cash, the OPET network and growth investments provide a real floor.
My objection is narrower: a good company and a cheap share are not the same sentence. At 267.75 TL, the investor is not buying a broken refinery at a discount. The investor is paying in advance for the midpoint of guidance, the eventual release of the derivatives block and an eight-year gross-EBITDA duration.
Which data holds the key to the treasury?
The trio that would pull the thesis lower is clear: realized net refinery margin falling below 6 dollars per barrel, period-end cash under the cash-flow statement definition falling below financial debt, and operating cash flow failing to cover profit even after collateral movements normalize. A continued decline in white-product yield and the 700 million dollar investment being financed with debt would accelerate the break.
The upside door is equally concrete: a margin above 7 dollars, at least 95% capacity, 30 million tonnes of sales, a substantial release of the 38.825 billion TL derivatives block, and post-investment cash exceeding the 22.8 billion TL case. Entek producing cash instead of losses, OPET returning to profit and SAF moving from the phrase “appropriate conditions” to commercial sales would convert options I assign no money to today into value.
Verdict: Fairly valued. More precisely, at the expensive edge of fair value. Tüpraş’s refinery is strong, its balance sheet is net cash, and its appetite for distribution is high; not all of the 700 million dollars of investment is maintenance spending. But the current price is not ignoring these virtues. It is already paying for eight years of EBITDA at the midpoint of guidance.
This share is not for an investor looking only for a straightforward dividend coupon. It suits someone who can stay cool when product margins, TMS 29 and derivative collateral become entangled in the same quarter. Owning part of Tüpraş means sharing not only in the barrel leaving the furnace, but also in that barrel’s second life in the contract book, and in the question of when the key to the treasury will return.