TD Cowen cut price targets across five consumer-finance stocks before third-quarter earnings, but the new targets describe very different businesses. Capital One and Affirm remained the firm's preferred names; Sezzle also kept a Buy rating, while SoFi and Klarna stayed at Hold. The useful investor question is not which target moved least. It is which earnings stream can absorb higher funding costs and credit risk while still compounding.
The source attributed targets of $17.50 for SoFi, $110 for Affirm, $242 for Capital One, $16 for Klarna and $160 for Sezzle. Those figures are analyst opinions with a 12-month horizon, not company guidance. They should be tested against operating metrics rather than treated as independent facts about fair value.
Balance-sheet lenders and transaction platforms carry different risks
Capital One earns spread income and bears credit risk across cards and banking. It reported [second-quarter net income](https://investor.capitalone.com/news-releases/news-release-details/capital-one-reports-second-quarter-2026-net-income-30-billion-or) of $3.0 billion, or $4.73 per diluted share, and released $662 million of loan-loss allowance, primarily because domestic-card performance was better than assumed. That release boosted the period; investors should separate recurring revenue from reserve movement.
SoFi combines held-for-investment lending with a fee-based Loan Platform Business. Its second-quarter release showed $143.3 million of adjusted net revenue from the platform, including $140.9 million tied to $3.1 billion of personal loans originated for or referred to third parties. That model can serve demand without putting every dollar on SoFi's balance sheet, but partner appetite and sale terms determine the economics.
Affirm, Klarna and Sezzle are more closely tied to payment volume, merchant relationships, funding and consumer repayment. Klarna's [Q2 release](https://investors.klarna.com/News--Events/news/news-details/2026/Klarna-Reports-Second-Quarter-2026-Results/default.aspx) reported $0.01 of EPS versus a $0.14 loss a year earlier, with credit provisions at 0.52% of gross merchandise value. Low loss ratios are encouraging, but rapid volume growth can mask seasoning until newer cohorts mature.
Upside percentages are not comparable without risk
The source calculated roughly 47% upside to Affirm's target and 11% for SoFi. A larger target gap may mean more expected growth, more perceived mispricing or simply higher uncertainty. Capital structure, stock volatility and estimate dispersion differ. Comparing the percentages without those dimensions turns a research note into a ranking it was not designed to provide.
The companies also recognize economics at different times. A card lender earns interest over a loan's life and reserves for losses. A platform may recognize fees near origination while retaining servicing duties. A buy-now-pay-later provider can turn receivables quickly but depends on funding and merchant fees. Revenue growth therefore does not map to cash generation in one uniform way.
Valuation duration differs as well. Capital One's current earnings and tangible capital can anchor analysis; younger platforms are valued more heavily on future scale and margin. When rates rise, distant expected cash flows usually suffer a larger present-value hit, even if near-term operating metrics remain sound.
The clean pre-earnings comparison uses four questions: Is transaction or loan growth organic? Are credit losses stable on a seasoned basis? Is funding cost falling or rising? How much incremental revenue becomes operating profit after incentives and marketing? Answers to those questions will matter more than retail sentiment or a one-day target revision.
TD Cowen's preference for Capital One and Affirm can be read as a barbell: a profitable diversified lender with substantial reserves, and a faster-growing commerce network with higher duration. SoFi offers a hybrid, while Klarna and Sezzle expose investors to different mixes of BNPL growth and execution. The target cuts do not establish a sector-wide deterioration; they tighten the valuation hurdle before results.
Third-quarter releases should be judged against company-specific drivers. For SoFi, partner originations and platform fees matter. For Capital One, card loss rates, funding and integration effects matter. For the BNPL names, GMV, transaction margin and delinquency trends matter. A disciplined comparison turns five target prices into five operating tests rather than one popularity contest.