OneMain Holdings: a great lender, priced for great credit

OneMain is one of the best-run nonprime lenders in the United States. It earns returns on equity above 20%, pays a 7.3% dividend and shrinks its share count every year. The problem is not the business. At almost twice book value, the price already assumes that the good credit years last.

At $57.25 on September 23, 2026, the shares trade 6.1% above my triangulated Base value of $53.96. The Bear value of $32.42 is 43.4% below the price, and the required margin of safety of 35% puts the maximum buy price at $35.07. For existing holders the dividend and buybacks justify holding. For new money it is a watchlist name.

Route: SF-02 Financial institutions, nonprime specialty lender without deposits. Economic classification: raw score 7/12, valued with a financial-sector standard engine because credit and funding cycles make long-duration models falsely precise. Primary method: justified price to book from sustainable ROE. Secondary methods: normalized earnings and historical price to book. Values are present fair-value estimates, not 12-month price targets.

What OneMain is

OneMain lends to nonprime American consumers through branches and digital channels. Personal loans are the core, with auto finance and credit cards growing around them. At June 30, 2026, managed receivables were $26.9 billion, up 7% from a year earlier, and the company passed 4 million customer accounts. Consumer and Insurance is its only reportable segment.

It is not a bank. It funds itself with $23.1 billion of debt, 52% of it secured, instead of deposits, and its corporate ratings are below investment grade (BB, Ba2, BB+). Bank metrics such as CET1 or loan-to-deposit ratios do not apply. Free cash flow is not a meaningful anchor either, because loans are the raw material of the business. The right anchor is return on common equity after actual charge-offs.

The character of the business is broadly unchanged. Cards and auto broaden the franchise without turning it into a different company, though fast card growth and whole-loan sales are worth watching.

The latest quarter

Management guides full-year C&I net charge-offs of 7.4% to 7.9% and expects significant improvement in the second half and into 2027, citing better early-stage delinquencies. Charge-offs are still higher than a year ago, so that improvement has to be delivered, not assumed.

Liquidity is strong: $567 million of cash, $1.0 billion of undrawn revolver, $6.5 billion of undrawn conduit and card funding capacity and $11.6 billion of unencumbered receivables.

Normalized earnings

Normalized EPS is set at $6.50: above the five-point average because the share count keeps falling, but at the current run rate rather than above it. Book value per share is about $29.40 ($3.383 billion of equity), so the stock trades at 1.95x book. Sustainable ROE in the Base case is 21%, the midpoint of a 20% to 22% range. Book value growth after distributions is assumed at 3% a year.

Primary valuation: justified price to book

For a lender, fair price to book = (ROE minus growth) / (cost of equity minus growth).

All values use the current book value of $29.40. The primary Base value is $52.92.

Secondary valuation: normalized earnings

Normalized EPS of $6.50 at 8.5x gives $55.25. The Bear case uses EPS of $4.50, close to the 2024 trough and consistent with ROE of about 15%, at 7.5x, giving $33.75. The Bull case uses $7.25 at 10x, giving $72.50. Each scenario uses earnings consistent with its own credit assumption.

Historical cross-check

OneMain traded near 1.0x book in late 2022, around 1.5x to 1.7x through much of 2024 and around 1.8x to 2.2x in 2025 and 2026. Using 1.2x, 1.85x and 2.2x for Bear, Base and Bull gives $35.28, $54.39 and $64.68. Today's 1.95x is defensible while ROE stays above 20%, but it sits well above the valuations the market paid in stressed credit years.

Valuation triangulation

The primary and secondary methods diverge by only 4.3%, and the same weights apply in all three scenarios. The result is not rounded up: every method on its own gives less than the current price.

What is the market already pricing in?

At 1.95x book, with 3% growth and a 13% cost of equity, the price implies a sustainable ROE of about 22.5%, above the 21% Base. The market is slightly more optimistic than this analysis. That leaves little room for a credit disappointment: the price is fair only if charge-offs improve as guided and stay there.

Bear, base and bull scenarios

Bear: charge-offs stay elevated, ROE falls toward the cost of equity and the market pays close to book value, as it did in late 2022. Base: credit normalizes gradually along guidance and ROE settles around 21%. Bull: benign credit, ROE of 24% to 25% and a lower cost of equity.

The probability-weighted value is $50.74, 11.4% below the price. The asymmetry points the wrong way: the downside to Bear is almost twice the upside to Bull.

Price to book sensitivity

How the buy price is set

Confidence B normally calls for a 25% to 35% margin of safety. The top of that range is used because of nonprime credit cyclicality, structural leverage and below-investment-grade funding. A 35% discount to the $53.96 Base gives a maximum buy price of $35.07. A 45% discount gives a strong-buy reference of about $29.68. At $35.07 the current dividend would yield about 12%.

Dividend

The quarterly dividend is $1.05, or $4.20 a year, a 7.34% yield at $57.25. On normalized EPS of $6.50 the payout is about 65%, and buybacks come on top: diluted shares fell from 119.4 million to 115.8 million in a year. The dividend is covered today, but it is set by the board and depends on the credit cycle.

Reasons to own OMF

• High economic returns. ROE comfortably exceeds the cost of equity through a normal cycle.
• Scale in nonprime lending. Branches, digital channels and decades of underwriting data are hard to replicate.
• Cash returns. A 7.3% dividend plus net buybacks steadily increase value per share.
• Growth options and strong liquidity. Auto and cards add growth, and funding capacity is deep and diversified.

What could go wrong?

• Credit. Charge-offs sustained above 9% to 10% with rising delinquencies would compress ROE and book value growth.
• Funding. Below-investment-grade ratings make the company sensitive to credit spreads and capital markets access.
• Regulation. Consumer finance rules can affect pricing, collections and products.
• Pro-cyclical capital returns. High dividends and buybacks become a burden in a severe loss cycle.

Management execution

Funding and liquidity have stayed robust, the share count keeps falling and capital generation held up through the recent credit cycle, including the 2022 underwriting tightening that now shows up as a much better performing front book. The main open test is delivering the promised charge-off improvement. Management credibility scores 14/20, good.

Stock Analyza scorecard

The valuation score is 53/100: the methods converge well, but the price sits slightly above Base and far above the Bear value.

Confidence score

What I would watch from here

The next review point is Q3 2026 results and the 10-Q, where the promised second-half charge-off improvement should start to show.

The three most important thesis breakers

1. Net charge-offs sustained above 9.5% with rising delinquencies.
2. A rating or funding shock.
3. Sustainable ROE falling below 15%.

Adversarial review

The first draft of this analysis set the Base at $58, a Bull at $75 and a maximum buy price of $37.70. The recalculation lowered all three. The justified price to book ranges in the draft were overstated: at the draft's own inputs, ROE of 20% to 22% gives 1.7x to 1.9x book, which is $50 to $56, not $52 to $59, and the Bull inputs give about $69 to $72, not $72 to $78. The draft also lifted the mechanically weighted result of $55.83 to $58, above every individual method, which the framework does not allow. The Bear earnings value used EPS inconsistent with the Bear ROE. After correction, the verdict is unchanged, but the Base moves from slightly above the price to slightly below it.

The strongest counter-argument is that recent ROE already exceeds 22%, delinquencies are improving and management expects charge-offs to fall. If ROE stays at 23% to 25% while book value compounds, 2.2x to 2.4x book supports the mid $60s to about $70. That case is credible, but it depends on benign credit and funding, and today's price already pays for most of it.

Even the most generous single change keeps the Base at or just above the price, and none comes close to a buy price above $40. The publication gate status is PASS_WITH_WARNING: forward ROE, credit losses and the cost of equity are estimates and depend on the cycle.

Final verdict: hold at $57.25, watchlist for new money

OneMain is not expensive for what it is. But the price already capitalizes most of its high-return economics and offers little protection against a new nonprime credit downturn.

Verdict: HOLD at $57.25, WATCHLIST for new money. Triangulated Base value $53.96, downside 5.7%, Confidence B (19/25). Required margin of safety 35%, maximum buy price $35.07.

For an existing holder, the dividend and buybacks support holding. For new money, the risk and reward become compelling in the mid $30s, with stronger asymmetry near $30, provided the fundamentals remain intact.

The main thesis breaker is a simultaneous deterioration in delinquencies, charge-offs and funding costs that pushes sustainable ROE toward the cost of equity.

Sources

Primary sources include OneMain's June 2026 Form 10-Q, the Q2 2026 earnings release, the 2025 Form 10-K, the investor relations dividend history and quarterly results, and the Q2 2026 earnings call for charge-off guidance. Historical price to book comes from Macrotrends and serves as a secondary input.

The reference market price is $57.25, the regular session close on September 23, 2026. Balance sheet figures are as of June 30, 2026.