Navient Corporation NAVI

Key data

The business

Navient owns a shrinking pile of student loans it no longer adds to. Two kinds matter. The old federal loans, made under a program the government backstops on nearly every dollar, throw off thin margin and melt a little each quarter. The private education loans, its own credit at its own risk, carry the real margin and the only losses large enough to move the company. A third leg, business processing for hospitals and government agencies, is fee income that grows but stays small. The engine is not growth, it is runoff: collect principal and interest as the loans amortize, fund them in the securitization market, and hand the cash back through the dividend and buybacks. There is no moat here, only a liquidation run by a servicer cheap enough to work its own legacy book. The company left federal loan servicing entirely after a 2024 regulator settlement, and now runs on 670 people. The thing a customer actually holds is a monthly private loan payment on a degree bought a decade ago.

The numbers

Read the runoff top down: the interest engine is cooling while the share count falls faster, so per-share figures hold up better than the totals.

The quarters show the shape of the risk more than the trend. Four of the last five earned money at a modest, steady clip, and one wiped out more than the other four made combined on a single charge. Strip that quarter and this is a small, boring, positive earner. Leave it in and you see that a runoff sold as predictable is not.

The interest income has more than halved in four years, and that is the business doing exactly what a runoff does. The offset is the share count, down about a third over the same stretch, which is why half-year earnings are back in the black even as the book keeps shrinking. Cash generation looks enormous next to the price, close to half the market value in a single year, but almost all of it is the portfolio converting itself into cash, not a stream that repeats. What this memo believes that the tape does not: the pile pays out more than $9.63 before it disappears, and the one figure that settles it is the private loan loss rate.

Management

The record here is written in the buyback column, and it cuts both ways. The company retired roughly a third of its shares over five years, and every year's average purchase price sits above today's quote, so on a market-value basis all of it is underwater. On the company's own carried value it is the opposite: each of those purchases was made below stated worth, so per-share value rose even as the price fell. The pace is slowing as the cash engine shrinks, from $600M a year to a ninth of that. The dividend has held at $0.64, a 6.6% yield covered several times over by cash flow. Insiders neither bought nor sold in the open market over the last year, which tells you nothing. What tells you something is the activist Edward Bramson, who runs Sherborne and sits on the board, pushing liquidate-and-return over reinvention.

How it fails or surprises you

The charge that shouldn't have happened. One quarter in the last five erased more than the other four earned, with little warning, on a book sold as predictable. That is the fact this read explains least well. If such hits recur, reserve builds, loss on a loan sale, or a fresh regulatory bill after the 2024 ban, the return-of-cash case leaks a bucket at a time. Watch quarterly provisions and any new legal reserve.

The engine cools faster than the count falls. Interest income keeps stepping down each quarter while funding costs stay high. If the book runs off faster than modeled, or the securitization market tightens, run-rate earnings can slip back toward zero before buybacks shrink the share count enough to hold per-share results up. Watch sequential net interest income and the margin on the remaining loans.

Paid for a fraction of what it may pay (right tail). At 0.46x tangible book the market prices a bad ending. If the private loans pay near par and the guaranteed loans wind down clean, the runoff returns well above $9.63, and one ordinary year of profit plus the 6.6% dividend pulls the price toward stated value. Watch the private loss rate holding near current levels and book value per share firming as shares fall.

Closing thoughts

This is an uncertainty a single print resolves, sitting on top of an exposure only survivability answers. The whole question is how much cash comes out of a melting book and how cleanly, and the number that governs it is the private education loss rate. The left tail is concentrated: the guaranteed loans are backstopped, so the danger lives in the private book and in surprise charges like the one that already hit. I judge the downside better-defined than the upside is large. The value at half of tangible worth is real if the loans pay, but the shrinking interest income and that one lost quarter say the runoff is not smooth, and this is the kind of book that looks like free money at half of stated value until a credit cycle reminds you why it trades there.

The bet is still that the old student loans pay back more cash than the whole company costs today, that the private borrowers keep paying and the guaranteed loans wind down clean. What breaks it is a private loss rate that climbs and an interest engine that turns run-rate earnings negative before the share count falls enough to matter. The pair that tells you first is the private education charge-off rate and the quarterly net interest income. Nothing has printed since the last read, when that charge-off rate near 1.84% and the loan-recovery trend both still pointed the right way, so the coming quarter is the one that retests it, and if that loss rate pushes through roughly 3% while interest income keeps sliding, the runoff returns less than you paid and the thesis is wrong.