Bezeq in the first quarter: cash flow jumped, but the payout test is just starting

Bezeq opened 2026 with a quarter that strengthens the base case, but does not remove the frictions around it. Free cash flow jumped to NIS 455 million, core revenues crossed NIS 2 billion, fiber is still lifting ARPU, and yes has now posted a quarter with positive profit and cash flow. Still, this is not a clean victory lap: part of the cash-flow jump came from a tax refund and working-capital movement, reported profit was hit by a one-off Pelephone labor expense, and the company has already approved a NIS 549 million dividend plus a buyback plan of up to NIS 150 million. The quarter therefore does not prove that Bezeq can keep increasing distributions without effort. It sets a more precise test: preserve core growth and AA-level credit ratings while 2026 still includes employee retirements, heavier working capital, relatively high CAPEX, and strategic options that have not yet become binding agreements. What worked this quarter was the transition from infrastructure that consumed investment to infrastructure that is beginning to generate more revenue per customer. What is still missing is proof that all-in cash flexibility after dividends, buybacks and investment remains comfortable without help from quarterly tax timing.

The Quarter Is Stable Only After Cleaning The Noise

The accounting headline is weaker than the underlying activity. Net profit fell to NIS 211 million from NIS 303 million in the corresponding quarter, mainly because of expenses related to Pelephone's renewed collective agreement. But comparable net profit, which excludes exceptional effects and valuation movements, rose to NIS 300 million from NIS 288 million. Adjusted EBITDA rose to NIS 928 million, and adjusted net profit was NIS 307 million.

That gap matters because 2026 is supposed to be a stabilization year. The company left its full-year outlook unchanged: adjusted EBITDA of NIS 3.7 to 3.8 billion, adjusted net profit of NIS 1.0 to 1.1 billion, and CAPEX of about NIS 1.6 billion. This is a base-preservation outlook, not a jump. Following the previous annual analysis, the quarter gives a partially positive answer to whether investment in fiber and yes is beginning to convert into cleaner profit and cash flow.

Group core revenues grew 2.6% to NIS 2.03 billion and represented about 94% of total revenues. That is stronger than the reported total revenue line, which declined to NIS 2.169 billion because of non-core components, including Pelephone interconnect and the deconsolidation of Bezeq Online. This is what Bezeq's move from the old world to the new world looks like: the reported top line is less impressive, but the line management wants the market to measure is still growing.

Pelephone is the cleanest example of the noise. Revenues excluding interconnect slipped to NIS 518 million, and adjusted EBITDA fell to NIS 178 million, but the pressure came from lower roaming revenues due to the war with Iran. The company says service revenues would have grown by about 2% after adjusting for that impact. At the same time, Pelephone added 34 thousand postpaid subscribers, its highest quarterly growth since the second quarter of 2018, and 5G postpaid plans reached 1.45 million, or 61% of postpaid subscribers.

Fiber And Yes Are Giving Better Proof

In the fixed-line business, the question is no longer whether the fiber network is large enough. It is. At quarter-end, 1.029 million subscribers were connected to the fiber network, and by the reporting date the number had risen to about 1.04 million. Deployment reached about 2.99 million homes passed, and take-up reached about 35%. The current question is monetization quality.

Here the quarter is positive, but not perfect. Fixed-line core revenues rose 2.5% to NIS 997 million, while telephony revenues declined to NIS 113 million. Retail broadband ARPU rose to NIS 139 from NIS 134 in the corresponding quarter, another step toward the 2029 target of about NIS 150. But retail broadband lines declined to 968 thousand, while wholesale lines rose to 504 thousand. Fiber is filling the network, but part of the growth is still coming through wholesale rather than direct retail customers.

yes provides the newer proof point. Revenues rose 7.5% to NIS 343 million, adjusted EBITDA rose 14% to NIS 56 million, and free cash flow was NIS 21 million. After years in which yes was mainly associated with impairments, this quarter shows a business that is already contributing operationally. TV subscribers remained 565 thousand, but IP subscribers reached 499 thousand at quarter-end and 503 thousand near the reporting date, or 89% of total subscribers. Fiber subscribers at yes rose to 132 thousand at quarter-end and about 137 thousand after the date.

Still, the quality of yes growth is not the same as a simple recovery in TV pricing. yes ARPU rose to NIS 202, but the increase was driven mainly by TV-plus-fiber bundles and revenues from the Partner transaction. That is positive because it proves that the company is building new revenue layers. It also means that investors looking only for a recovery in classic TV pricing may flatten the read. yes is improving because the product is changing, not because the old product suddenly regained full pricing power.

The international and business services activity tells a similar story. Revenues rose to NIS 289 million, and net profit rose to NIS 15 million. But a NIS 14 million impairment was still recorded in the background, and the activity's DCF value remained negative at NIS 93 million. This is a useful reminder of the difference between a good operating quarter and a full asset-value repair.

Cash Flow Jumped, But The All-In Cash Picture Is Less Simple

The company's free cash flow is defined as operating cash flow less net payments for fixed assets and intangible assets, and less lease payments. This is an operating cash view after CAPEX and leases, but before dividends, buybacks and debt repayments. In the first quarter, that metric was very strong: operating cash flow of NIS 1.007 billion, less NIS 431 million of net investments and NIS 121 million of lease payments, produced free cash flow of NIS 455 million.

That number matters, but it does not tell the whole story alone. The increase in operating cash flow came from lower income tax paid in the fixed-line activity and working-capital changes, together with improvement in the core businesses. In other words, there is operating progress, but also timing. It would be wrong to take NIS 455 million and multiply it by four without testing the source.

The second layer is all-in cash flexibility. After the balance-sheet date, a NIS 549 million dividend was approved for payment on May 14, 2026, and a buyback plan of up to NIS 150 million began. By the approval date, about NIS 70 million had already been executed, roughly 46% of the plan. The company has therefore already committed to a meaningful shareholder-return path in a year that still includes investment, streamlining and working-capital needs.

The important external signal comes from the rating agencies. Midroog left its Aa2.il rating with a stable outlook, and Maalot affirmed ilAA with a stable outlook. That supports the view that this is not a financing-stress story. But Midroog also assumes that, on an all-in cash view after working capital, investments and dividends, the company will generate negative free cash flow of about NIS 350 to 400 million in 2026, before moving to positive cash flow of around NIS 400 million in 2027. It also expects adjusted net financial debt of NIS 7.2 to 7.5 billion during the forecast years, versus NIS 6.93 billion at the end of 2025.

This is not a sharp credit warning. The ratios are still expected to fit the rating level. But it is a reminder that the higher distribution is not funded by cash with no competing use. It sits on a strong company with good financing access, but also one with CAPEX, leases, working capital and retirement programs that still consume cash.

What The Market Will Test Next

The first trigger is cash-flow quality. If free cash flow stays high in the coming quarters without relying on tax and working-capital timing, the payout case will strengthen. If the first quarter proves unusual, the market will return to the gap between company-defined free cash flow and all-in cash flexibility after dividends and buybacks.

The second trigger is yes. A quarter with positive profit and cash flow is a milestone, not the end of the test. The company needs to show that contributions from fiber bundles, Partner and the IP migration repeat through the year, not just in one quarter. If yes ARPU keeps rising and fiber subscribers keep growing without erosion in the subscriber base, it will be easier to treat the improvement as operating progress rather than a valuation model.

The third trigger is Pelephone. The decline in ARPU to NIS 44 this quarter is explained by lower roaming due to the war with Iran, so it looks more temporary than structural. But the NIS 86 million retirement expense and the new collective agreement shift the test to the next few years: whether streamlining really offsets employee benefits, and whether 5G and 5G MAX growth improves service revenues once roaming normalizes.

The fourth trigger is the submarine cable. The March 24, 2026 memorandum includes an expected group investment of about NIS 250 million over three years, with an intention to establish an international cable system between East and West through Israel. This could become an interesting strategic engine, especially because it also relies on a terrestrial segment using the group's fiber infrastructure. But there is no detailed agreement yet, no certainty of completion, and the parties are expected to formulate an agreement within six months from signing. For now, it is optional upside, not a 2026 earnings anchor.

Above all of this sits structural separation. The 2026 and 2029 targets do not include its cancellation, so Bezeq's operating base stands even without regulatory change. But any concrete progress could change how the market values the group's ability to capture value from yes, fiber and bundled services.

The first-quarter conclusion is that Bezeq reads more like a cash-return equity under credit discipline than a fast-growth story. That profile can work well as long as cash flow repeats, yes keeps contributing, and shareholder returns do not push debt beyond the rating scenario. If the next quarters show that the first quarter leaned too heavily on tax and working-capital timing, the report will look like a favorable window rather than a step change. If cash stays strong and yes improvement repeats, 2026 will look more like the first harvest year from the investments made in recent years.