Samsung SDI Co., Ltd. (KRX:006400) made ₩203.8bn of operating profit in the June quarter, its first after six losses. An earlier piece here questioned the quality of that profit, mainly because of the tax line. This one goes a step further down the statements, to the line that decides whether a turnaround is real: free cash flow.
Operating cash flow in the June quarter was ₩24.2bn. Purchases of plant and equipment were ₩534.5bn. Free cash flow, on the simplest definition, was negative ₩510.3bn. That is the eighth consecutive quarter below zero.
My view is that the company has fixed its income statement and not yet its cash flow statement, and that the ESS story it tells about the first is the thing to test against the second.
Here is the record from DART, oldest first, for the eight quarters from Q3 2024 to Q2 2026. Operating cash flow ran ₩111.3bn, ₩375.7bn and ₩172.3bn. It then ran ₩529.3bn, negative ₩241.4bn and ₩332.2bn. The final figures were ₩85.2bn and ₩24.2bn. Plant purchases ran ₩1,251.4bn and ₩2,193.5bn. They then ran ₩874.0bn, ₩997.5bn and ₩539.0bn. The final figures were ₩656.4bn, ₩591.4bn and ₩534.5bn.
Subtract one from the other and every quarter is negative. The shortfalls started at ₩1,140.1bn and ₩1,817.8bn. They then ran ₩701.7bn, ₩468.2bn and ₩780.4bn. The final shortfalls were ₩324.2bn, ₩506.2bn and ₩510.3bn. The total is ₩6,248.9bn of cash the business consumed over two years beyond what it generated.
The shape has changed, which matters. Capital spending fell from ₩2.19tn in the last quarter of 2024 to around ₩550bn a quarter now. Annual plant purchases were ₩6.27tn in FY2024, ₩3.07tn in FY2025 and ₩1.13tn in the first half of 2026. The company has cut its build rate by more than half. What hasn't happened is the other side. Operating cash flow was ₩85.2bn in March and ₩24.2bn in June. The quarter that returned to profit produced almost no cash.
The income statement and the cash flow statement disagree by about ₩450bn in the June quarter. Net income was ₩471.6bn. Operating cash was ₩24.2bn. The difference sits in working capital.
Inventories rose to ₩3,675.5bn at June 30 from ₩3,330.1bn in March and ₩2,936.3bn at year end. That is a 25% increase in six months. Trade receivables rose to ₩3,742.8bn from ₩3,125.8bn in March, up 20% in a single quarter. Together, stock and receivables absorbed roughly ₩960bn more cash in the first half than they held at the start of the year.
Some of that is what a recovery looks like. A battery maker ramping production for new energy storage contracts builds cells before it ships them and ships them before it is paid. But the company hasn't said that. The quarterly statements show the numbers without the explanation, and I'd rather flag the gap than fill it.
Cash and equivalents were ₩1,505.5bn at June 30. A year earlier they were ₩2,154.4bn. At the end of FY2024 they were ₩1,885.1bn. The balance has fallen even though the company raised a great deal of money.
The funding came in three forms. In the June quarter of 2025, financing activities brought in ₩1,975.9bn. Share capital rose from ₩356.7bn to ₩415.8bn. That is a rights issue. In the first half of 2026 financing brought in a further ₩637.7bn. That was ₩329.9bn in March and ₩307.8bn in June. It is borrowing. And current liabilities jumped to ₩12,649.8bn at June 30 from ₩10,305.0bn in March. That is a ₩2.3tn increase in one quarter. Some of it is long-term debt moving to short-term. Some is payables.
Total liabilities were ₩20,785.7bn against equity of ₩26,836.2bn. That is not a stretched balance sheet. But the equity figure has its own story. It rose ₩1,926.3bn in the June quarter on ₩471.6bn of net income. That was because other comprehensive income was ₩1,903.7bn. That looks like the market value of an investment holding, and it can move the other way.
The company's explanation for the profit turn is energy storage. The first-half release says high-output products drove results. These were products for power-grid ESS, uninterruptible power supplies and battery backup units for AI data centres. It says the company signed long-term supply contracts with major US ESS customers and won a Korean next-generation distribution-grid ESS project. In March, management told Korean press it had ESS volumes secured for three to four years. It said it would reach fully non-Chinese US supply next year.
The company does not say what share of revenue ESS now represents. Neither quarterly release nor the DART statements give that share. Battery division revenue was ₩3,519.0bn in the June quarter and battery operating income was ₩159.3bn, a 4.5% margin. By subtraction, the electronic materials division made ₩249.8bn of revenue and ₩44.5bn of operating income. That is a 17.8% margin. That is as far as the disclosure goes. ESS, electric vehicle and small cells are all inside the battery line.
Two earlier pieces here covered the Ulsan plan for cheaper battery chemistries and the buyout of General Motors' stake in the Indiana plant. The company is turning that plant toward ESS. Both are ESS bets. Neither comes with a revenue figure for the business they are betting on.
That matters for the cash question. ESS contracts with US utilities and data-centre operators tend to be long-dated and volume-committed. That is good for revenue visibility and bad for working capital. The customer's payment terms are usually long. If ESS is half of battery revenue, the receivables build is a feature of the new business and will stabilise. If ESS is a fifth and the rest is still electric vehicle cells sold into a weak market, the build is inventory that isn't moving.
The company's market capitalisation was ₩44.2 trillion at Friday's close of ₩548,000. Book equity was ₩26,836.2bn, so the shares trade at 1.6 times book. Trailing four-quarter net income is ₩325.6bn, so the multiple is above 130 times. The shares are 23% below the 52-week high of ₩712,000 and 2.8 times the low of ₩198,100.
The market is not paying for the June quarter. It is paying for what the ESS business could earn once the Indiana plant and the US LFP line are running. At 1.6 times a book that includes ₩20.3tn of plant, the price assumes that plant earns a return. Eight quarters of negative free cash flow say it hasn't yet.
The bear reading of the cash flow could be a timing error. Battery makers collect unevenly, and a strong fourth quarter of collections would reverse most of the first-half working capital build. FY2025 operating cash was ₩792.4bn for the year despite two weak quarters inside it.
The capex cut could also be more than a pause. If the company holds plant purchases near ₩550bn a quarter and operating cash recovers to the ₩500bn level it reached in mid-2025, free cash flow turns positive. That would happen without any heroics on the revenue line.
And the ESS share could be large. Management's confidence about multi-year volumes and its decision to convert an EV plant to grid batteries both suggest a business that is already substantial. The company may simply not want to tell its EV customers how much it now depends on something else.
The third-quarter report in late October has two lines. Operating cash flow above ₩300bn, with inventories flat or down, would say the June quarter's working-capital build was a ramp. It would mean the cash is now following the profit. Operating cash flow below ₩100bn again, with inventories rising, would be a ninth negative quarter of free cash. It would mean the balance sheet is still paying for the turnaround.
The second thing is any disclosure of ESS revenue share. The company has managed two quarters of results calls without giving one. A number above 40% of battery revenue would explain the receivables and the confidence. Continued silence would say the number isn't one the company wants to show.
kstock reads DART every morning and writes up what moved — the contract, the buyback, the number that does not add up. The daily post and a Saturday roundup, by email.