The forward multiple trading above the trailing multiple is the key tell: with no forward P/E available and a wide guidance range, the market appears to be pricing a denominator that has not yet stabilized rather than one that is growing. P/B at 1.00 versus Delta's 2.49 and Southwest's 3.00 suggests investors are assigning essentially no franchise premium to the Seattle and Portland hub position or the Mileage Plan, which may reflect either genuine doubt about return durability or a de-rating driven by the elevated 2.07 debt-to-equity. EV/EBITDA of 9.25 looks optically cheap against LUV at 12.13 and AAL at 12.31, but that comparison warrants investigation given EBITDA volatility evident in the D/EBITDA swings, so the trailing multiple may be flattered by a quarter in which reported EBITDA was abnormally high.

The most reliable measure of true earning power here is operating margin before below-the-line items, because the reported $147M net loss against $82M of operating income implies roughly $229M of charges that sit outside operations and will not recur at the same magnitude. Stripping those, the underlying picture is still one of erosion: operating income fell 76% to $82M over the same two-year window while SG&A rose 44%, suggesting overhead scaling is not being governed by the top line. Gross margin readings ranging from 16.3% to 100% across the ten quarters are too volatile to interpret as pricing power, so investors should anchor on operating margin and CASM-ex trends instead, which at 1.9% versus a 7.5% print in 2025Q2 indicate limited cushion against further fuel moves.

The decay is margin-driven rather than efficiency-driven: asset turnover has been essentially flat in a 0.16–0.20 band across all ten quarters, meaning the balance sheet is being worked at a constant intensity while the return extracted from it collapses. This matters because capex of $3.75B over the period against $1.88B of depreciation means invested capital is growing well beyond replacement needs, so the numerator must rise to justify the denominator — and it is falling instead. On that pattern the company appears to be compounding capital at returns well below any conventional cost of capital estimate, which would imply the ongoing fleet and acquisition investment is dilutive to value unless RASM recovers materially; the divergence between cumulative operating cash flow of $3.18B and cumulative net income of $211M does not change this, since depreciation explains the gap rather than resolving the return shortfall.

Total debt rising from $3.9B to $7.3B alongside equity declining from $4.5B to $3.5B means the leverage shift is a two-sided event, and the coverage trajectory suggests the cushion that once existed has largely been consumed. D/EBITDA of 24.49 in 2026Q3 versus 8.61 in 2024Q2 is extreme, though the 195.28 print in 2026Q2 signals the denominator was near zero rather than a step-change in borrowing, so analysts should treat the series as reflecting earnings volatility as much as debt growth. Compared with Delta at 1.02 and broadly in line with United at 2.03, ALK now carries legacy-carrier leverage without legacy-carrier margins, which investors should monitor closely given the -0.63 and -3.92 coverage prints that occurred in quarters of negative operating income.

Because days inventory sits at only 7, the quick and current ratios are nearly identical, meaning liquidity here is essentially a cash-and-receivables question with no inventory liquidation value to lean on in stress. The 0.46 level would be conventional for an airline carrying deferred ticket revenue as a current liability, so the analytical signal is not the absolute number but the two-year downtrend concurrent with a -6.8% FCF margin in 2026Q3 and $1.11B of buybacks. Under a demand shock comparable to a West Coast tech-driven travel pullback, this configuration suggests the company would depend heavily on operating cash generation rather than balance sheet slack, and given the -3.4% net margin, investors should watch whether cash balances continue to erode before judging the buffer as adequate.