Self-Help

Self-Help

Molina can outlast its cost-trend problem: no debt matures before 2028, a $1.25 billion revolver sits undrawn, and corporate leverage runs at ~48% of capital against a 60% covenant. It bought back $1.0 billion of stock in each of 2024 and 2025 — cutting the share count 8% last year — and has $500 million of authorization left. Two frictions temper the flywheel: management ranks repurchases third behind organic growth and acquisitions, and the freely-deployable cash sits at the parent, not on the consolidated balance sheet.

The balance sheet against the problem's duration

The dislocation here is a one-to-two-year medical-cost-trend miss that reprices through the 2026–2027 rate cycle (Dislocation, Damage Math). The relevant question for capital allocation is whether debt maturities or covenants force cash toward the lenders during exactly those years. They do not.

Source: FY2025 Annual Report (Form 10-K), Note 11 "Debt," contractual maturities table [1].

The ladder is deliberately long-dated. Nothing comes due in 2026 or 2027; the first maturity is the $800 million 4.375% notes in 2028, then $650 million in 2030, with the remaining $2,350 million spread across 2031–2033 [2]. Total principal is $3,800 million ($3,766 million carrying, net of issuance costs), and management puts the weighted-average life at roughly five years and the weighted-average coupon at 5.0%. As debt matures, Molina "typically engage[s] in a new private offering of debt to retire and replace the prior issuance" — the November 2025 issuance of $850 million of 6.500% notes due 2031, used to repay $740 million of term-loan borrowings, is the pattern in action [3].

Sources: FY2025 10-K balance sheet and Note 11 "Debt" [4]; covenant terms, Note 11 [5]. Debt/capital = $3,766M debt ÷ ($3,766M + $4,069M equity); interest coverage = $781M FY2025 operating income ÷ $192M interest expense.

Liquidity is ample on the numbers that bind. The $1.25 billion revolving facility was fully undrawn at year-end and runs to November 2030 [6]. Corporate debt-to-capital sits at ~48%, below the 60% covenant ceiling, and FY2025 operating income of $781 million covered $192 million of interest ~4.1 times even in the trough year [7].

Two facts cut the other way and belong in the same breath. First, debt rose $843 million in 2025 (from $2,923 million to $3,766 million) — some of that funded the buyback rather than operations, which is a choice to lever into the repurchase, not a forced paydown, but it is leverage added at a weak-earnings moment. Second, on February 4, 2026 Molina amended the credit agreement to temporarily reduce the minimum interest-coverage covenant from 3.0x to 1.75x for the quarters through December 2026, stepping back up to 3.0x by late 2027 [8]. That amendment is management creating covenant headroom ahead of a depressed-earnings 2026 — prudent, but also a signal of how thin coverage could get. It bites only if the revolver is drawn; at year-end it was not.

Where the cash actually is

For a managed-care insurer, the consolidated balance sheet flatters the self-help case. Molina held $4,248 million of cash and $4,008 million of short-term investments — about $8.3 billion — but most of that is regulated statutory capital held at the health-plan subsidiaries, which need roughly $3.1 billion of minimum capital and can only dividend excess to the parent with regulatory notice [9]. The cash that can actually buy back stock lives at the unregulated parent — and there was only $223 million there at year-end, down from $445 million [10].

The buyback fuel is therefore a flow, not the headline balance: the health plans dividended $985 million up to the parent in 2025 (and $997 million in 2024), against which the parent funded $439 million of capital contributions back down to the plans, $192 million of interest, and the $350 million ConnectiCare purchase [11]. The CFO frames the number that matters plainly: parent cash was "a little over $200 million" in Q1 2026 and is expected to exceed $600 million by year-end on continued upstreamed dividends — "that's where we can actually use it to redeploy" [12]. The reference line: Molina can comfortably outlast a two-year cost-trend problem — no maturity wall, undrawn revolver, an intact dividend-upstream engine — but the buyback capacity is governed by that ~$1–1.5 billion annual parent flow, not by the $8 billion of cash a screen would show.

The repurchase record — executed, not just authorized

This is a company that repurchases, and the framework's hard-fail — a share count rising on stock-based comp or serial acquisition dilution — does not apply. Cash actually spent, straight from the cash-flow statements:

Source: derived from reported cash-flow statements, FY2019–FY2025 10-Ks; matches fit_features.share_count_trend.buyback_cash_per_year. FY2024–FY2025 detail from Note 13 "Stockholders' Equity" [13].

The tempo picked up as the stock fell. In 2025 Molina spent $1.0 billion: ~1,679,000 shares in Q1 at an average cost of $297.83, and ~2,849,000 shares in Q3 at an average cost of $175.50 [14]. The blended ~$220.85 sits fractionally below today's $224.82 — the Q1 tranche at $298 looks poor in hindsight, but the company leaned harder into weakness in Q3 at $175, buying 70% more shares for the same $500 million. The Q3 purchase came with an explicit valuation call from the CFO: "We see real value in our shares at current market prices, which we believe at this low point in the rate cycle underappreciate the longer-term margin targets of our business" [15].

The share count confirms real retirement, not optical churn:

Source: reported shares outstanding, FY2018–FY2025 10-Ks; matches fit_features.share_count_trend (5-yr CAGR −2.5%, rising = false).

Shares fell from 57.7 million (FY2024) to 52.9 million (FY2025) — an 8.3% reduction in a single year — and from 58.6 million five years ago, a 5-year CAGR of −2.5% [16]. (The 2018 peak of 66.6 million reflects convertible-note dilution long since worked off.) The authorization is live, not aspirational: an April 2025 board authorization of $1 billion runs through December 31, 2026, of which $500 million remained available as of February 10, 2026 [17].

Management's buyback intent, from the record

The brokers ask every quarter, and management's answer is consistent — and consistently ranks the buyback below other uses. Asked in Q2 2025 how repurchases fit against acquisitions, CEO Joseph Zubretsky was explicit: repurchases are "part of our capital strategy, but [rank] third in our priorities. The primary focus is on organic growth… The second priority is mergers and acquisitions, where we are acquiring companies just above their book value… we always have an eye towards share repurchase" [18]. In Q3 2025 he reaffirmed the ordering unchanged — "organic growth, inorganic growth, and returning capital to shareholders through share repurchase" — while noting Molina produces "$1.5 billion of capital capacity a year even at these compressed margins" [19]. The CFO sizes the choice as $1.5–2 billion of deployable "dry powder" over the coming year [20].

The tension for the flywheel is genuine and two-sided. On one side, Molina executed $1.0 billion of buybacks in the dislocation year, called its own stock cheap, and holds $500 million of remaining authorization. On the other, management's stated preference at maximum fear is to buy distressed local Medicaid plans "at or around book value" — which it argues is "as good or even better than winning a new contract" — and to fund organic growth first [21]. At the exact moment the framework wants repurchases to be the priority, they are the third-ranked use of capital. A material acquisition would compete directly with the buyback for that ~$1.5 billion annual flow.

Insider buying alongside is thin. Across SEC Form 4 activity, one director — Richard Zoretic — made an open-market purchase of 800 shares at $125.16 on February 11, 2026, the exact day the stock troughed at $122.65 — a genuine but symbolic ~$100,000. Otherwise the flow was routine equity grants and tax-withholding dispositions, alongside a handful of executive open-market sales (the Chief Legal Officer sold ~17,800 shares for ~$3.3 million at $186 in May 2026). There was no broad insider accumulation into the decline. (SEC Form 4 filings, as reported; no single-filing PDF page.)

The levered exception does not apply

The framework tolerates debt only where the adjusted yield is extreme (~25–40%) and paired with a demonstrated multi-year share-count halving. That path is moot here on two counts: the adjusted yield is not in that range (see Yield), and Molina is not a levered balance sheet in the framework's sense — corporate debt of $3,766 million sits against $4,248 million of cash, i.e. mechanically net cash on the parent's debt, with leverage of ~48% of capital and ~2.5x trailing EBITDA [22]. The self-help case rests on the ordinary buyback engine, not the levered exception.

The absurdity check

The deterministic float_retirement_years feature is not computable: adjusted FCF could not be derived (stock-based compensation is not separately captured in the financial feed for any year), and FY2025 reported free cash flow was negative at −$636 million — an outflow driven by the timing of Medicaid risk-corridor settlements and government receivables, not an earnings collapse (net income was still +$472 million), as the CFO underscored when he said parent cash flow, not consolidated operating cash flow, is what matters for a regulated insurer [23].

Illustratively, and not as a substitute for the feature: at the $11.9 billion market capitalization, consensus FY2026 free cash flow of ~$1.86 billion would retire the entire float in roughly 6.4 years; the ~$1.5 billion of annual capital capacity management cites implies closer to 8 years. Either figure is well outside the ~3-year "this cannot survive" zone the framework flags — the price is not making an arithmetic claim that a couple of years of cash flow could disprove. The honest version of the check is that it cannot be run on trailing adjusted FCF, and the forward illustration is ordinary, not absurd.

Dividend safety — not applicable

Molina pays no common dividend and states it intends to retain earnings to fund operations [24]. The return case leans entirely on the repurchase and the earnings recovery, not on a yield; dividend cover is therefore not part of this setup.

Management credibility

Zubretsky has run Molina since November 2017 — a long-tenured operator, not a serial promoter, and the acquisition and buyback commitments he has made have been delivered: eight acquisitions totaling roughly $11 billion of premium bought at ~22% of revenue, and $2.0 billion of repurchases across 2024–2025 [25]. That is not the promotional-CEO exclusion pattern of big claims with no follow-through.

The guidance record, however, is genuinely dented, and it is the sample that matters most here — three consecutive years of the year-opening EPS promise:

Sources: FY2023 result and FY2024 guide, Q4 FY2023 call [26]; FY2024 result, Q4 FY2024 call [27]; FY2025 guide and revised outlook, Q1 and Q3 FY2025 calls [28].

FY2023 beat: guided "at least $19.75," delivered $20.88 (+17%), squarely inside the then-stated 15–18% long-term target [29]. FY2024 missed: guided "at least $23.50," delivered $22.65, with management acknowledging "our full year results [fell] below our guidance" on fourth-quarter medical-cost pressure [30]. FY2025 missed severely: the year opened with guidance of "at least $24.50," reaffirmed in Q1, then cut to ~$19 and finally to ~$14 by Q3 as the cost-trend and Marketplace repricing hit [31].

Two readings coexist and the reader should hold both. The bear reading: back-to-back misses have hollowed out the "13–15% long-term growth" and "embedded earnings" framing the CEO has promoted, and insider ownership is low — all executive officers, directors and nominees hold 749,230 shares, just 1.44% of the company, so the skin in the game is modest [32]. The calmer reading: the FY2024 miss was modest and the FY2025 miss is the industry-wide medical-cost-trend forecasting error that is the dislocation the framework hunts — a repricing event, not a governance failure — and management met it by buying its own stock and disciplined acquisitions rather than by promotion. The instances above are the evidence; what would tip this toward the exclusion is a third guide-and-miss in 2026 with the stock talked up rather than repurchased.