Summary

  • Flex agreed to buy EPC Power for US$4.4 billion in cash, subject to the purchase agreement's adjustments, and arranged a 364-day senior unsecured bridge of up to the same amount.
  • Matching headline numbers do not make the bridge the final financing. The facility is conditional and available if permanent financing has not been obtained; Flex says it intends to replace it with debt and equity.
  • Acquisition close, financing takeout and the planned Q1 2027 separation of the Critical Power Infrastructure business are three separate clocks. The decisive missing record is the debt, equity, cash, ownership and covenant allocation of the future company.

The same number performs two different jobs

US$4.4 billion appears twice in Flex's new EPC Power transaction record. It is the headline cash consideration for all of EPC Power's equity. It is also the maximum size of a 364-day senior unsecured bridge commitment from Citi and Bank of America. The visual symmetry is tempting: a US$4.4 billion asset, a US$4.4 billion loan, one complete funding answer.

The contracts do not say that. Flex's 3 September Form 8-K describes the bridge as acquisition financing available if permanent financing has not been obtained, subject to the acquisition closing, the absence of a material adverse effect and other conditions. A commitment is not a draw. A ceiling is not a final debt balance. And 364 days is a temporary maturity, not evidence of the capital structure that the future independent business will carry.

The purchase price is not quite a fixed cash wire either. The equity purchase agreement starts the calculation at US$4.4 billion, then subtracts estimated leakage and estimated transaction expenses. It provides for closing statements and post-closing adjustment mechanics. The deal uses a locked-box enterprise value fixed as of 30 June 2026, with protection against leakage. This separates the agreed enterprise value from the final flow of cash at closing.

One number therefore anchors valuation, while the other limits a contingent source of temporary liquidity. Treating them as the same ledger would erase the contractual boundary that matters most.

Three clocks have to be kept separately

The first clock is completion of the purchase. Flex expects it in the fourth quarter of 2026, subject to closing conditions including Hart-Scott-Rodino clearance. The purchase agreement sets 1 October as an inside date. Its outside date is 31 December 2026, with two automatic three-month extensions in specified circumstances. None of those dates proves the conditions have been met or that funds have moved.

The second clock is permanent financing. Flex says in the transaction announcement that it expects a combination of debt and equity to replace the bridge. It does not disclose the size of either component, the price or buyer of an equity issue, the rate and maturity of new debt, security, covenants, rating consequences or the amount of existing cash that may enter the funds flow. The bridge buys execution certainty; it does not settle those choices.

The third clock is the planned separation of Flex's Critical Power Infrastructure business as an independent public company in the first quarter of 2027. EPC Power is meant to join that business. Closing the acquisition would put the asset inside Flex before the separation. It would not by itself determine which debt, cash, leases, working-capital facilities or liabilities follow the business into SpinCo, nor how much ownership Flex shareholders retain or receive.

This is the new information boundary. Flex had already announced the CPI separation before the EPC agreement. The September filing adds a large acquisition and a temporary funding backstop between announcement and separation. The question is no longer merely whether CPI will be separated. It is how an acquired business, a takeout financing and a new public balance sheet will be assembled in sequence.

The pre-deal balance sheet is context, not a financing answer

Flex's quarterly filing for the period ended 26 June 2026 shows US$2.840 billion of cash and US$5.219 billion of long-term debt before the EPC agreement. It also records US$1.134 billion of cash used for acquisitions during the quarter, US$2.830 billion of debt proceeds and US$1.385 billion of debt repayment. These figures show a balance sheet already moving through acquisitions and refinancing.

They do not permit a shortcut. Adding reported cash to the bridge ceiling would overstate available acquisition liquidity because the cash has operating, working-capital, jurisdictional and other claims against it. Subtracting cash from the purchase price would be equally speculative without the closing funds flow. Pre-deal long-term debt does not identify which obligations remain with Flex and which might be allocated to SpinCo.

The useful disclosure will be a sources-and-uses table tied to the close: final purchase consideration, fees, leakage and transaction-expense adjustments, cash used, debt issued, equity issued, bridge draw if any, and any subsequent repayment. A separate separation balance sheet should then show what moves to the independent company. Without both records, investors cannot distinguish acquisition financing from separation capitalisation.

Forecasts describe the asset being bought, not the capital required

Flex's press release forecasts approximately US$800 million of EPC Power revenue in 2026, about 40% organic growth in 2027 and an EBITDA margin of roughly 30% in 2027. The investor presentation also highlights more than 15 gigawatts deployed in 62 countries and anticipated manufacturing capacity above 30 gigawatts in 2027.

Those claims explain the strategic appetite. AI data-centre construction makes power conversion, grid interconnection and delivery timing economically scarce. A business that can convert utility power efficiently and ship at the pace of new campuses may command a large price because it sits close to the deployment bottleneck.

But forecast growth is not cash already available to service acquisition debt. A 2027 margin target is not an achieved covenant cushion. Deployed gigawatts do not state warranty exposure, customer concentration, backlog cancellation rights, working-capital intensity or the capital needed to expand factories. The forecasts should therefore be tested against actual revenue, gross-to-EBITDA conversion, cash conversion, backlog quality and capacity expenditure after closing.

The same discipline applies to synergies. Placing EPC Power inside CPI may broaden the product stack and customer access, but a forecast does not show which sales belong to EPC alone, which require Flex channels, or which integration costs arrive before revenue. Permanent financing should be sized against downside cash generation as well as management's central case.

The next filing should connect the ledgers

The most informative next record is not a repeat of the US$4.4 billion headline. It is a reconciliation across the three clocks.

At acquisition close, Flex should identify the adjusted consideration, fees and funding sources; whether the bridge was drawn; and the amount and terms of debt and equity used to replace it. After the close, reported EPC revenue, margin, cash conversion and capital spending should be separated clearly enough to test the forecasts. Before the separation, a pro forma SpinCo balance sheet should disclose cash, debt, leases, working-capital lines, pension or other material liabilities, allocation policy, related-party arrangements and ownership distribution.

Timing matters inside that sequence. If permanent financing is completed before closing, the bridge may never be used. If it is drawn, the period outstanding and takeout terms will reveal how much execution insurance cost. If market conditions delay the equity or debt issuance, Flex may face a different mix from the one envisaged at signing. If separation timing moves, the acquired cash flows may remain within Flex longer than expected.

None of these scenarios means the acquisition thesis is wrong. They show why a financing commitment and a final capital structure are different economic objects. The bridge transfers near-term closing risk away from the availability of capital. It does not transfer integration risk, forecast risk or the choice of who ultimately supplies that capital.

Flex has secured a route to the acquisition dock. Investors still need to see the permanent rails and the balance sheet that crosses the separation gate.

Sources