Caterpillar vs. Oshkosh: Which Industrials Stock Is a Better Buy in 2026?

Caterpillar (CAT) reported FY 2025 revenue of about $67.6B (+4.3% YoY) and net income around $8.9B, but net margin fell to 13.1% from 16.7% in FY 2024; free cash flow was $10.3B with debt-to-equity ~2.0x. Oshkosh (OSK) posted FY 2025 revenue of ~$10.4B (-2.9% YoY) with net income ~$647M (6.2% margin) and free cash flow of ~$618M, alongside lower leverage (debt-to-equity ~0.3x). The article frames CAT as higher-growth but more cyclical/high-valuation (forward P/E 38.8x vs OSK 13.1x) while OSK is positioned as a more defensive, government-contract-driven option (OSK ~20% of net sales from the U.S. government). Overall, it leans toward CAT for 2026 given ongoing AI/data-center related demand, despite valuation risk.

Analysis

The market is likely overpaying for the cleanest narrative here: CAT is increasingly a leveraged proxy for hyperscaler capex, grid buildout, and mining intensity, so the stock’s downside is less about one weak quarter and more about whether AI-related infrastructure spend stays linear. At this valuation, any evidence of data-center digestion, softer dealer inventory turns, or slower power-gen orders can trigger multiple compression before earnings actually roll over. The second-order tell is that if CAT’s end demand moderates, DE and PCAR usually feel it next via weaker replacement cycles and dealer caution.

OSK looks cheaper, but “defensive” is overstated. A meaningful slice of the earnings base is effectively a government timing bet, which means budget delays can create lumpiness in revenue recognition and working capital even when demand is intact. The lawsuit overhang matters less for near-term fundamentals than for how it caps the rerating multiple; this is the kind of name where one adverse legal or procurement headline can erase several quarters of perceived stability.

Over the next 1-3 months, the catalyst path is cleaner for CAT: order commentary, backlog mix, and margin guidance will decide whether the market continues to pay up for the AI-infrastructure thesis. Over 6-18 months, the real risk is that CAT’s growth story is structurally right but too expensive, while OSK’s low valuation is structurally warranted unless government contract visibility improves. The contrarian view is that the consensus treats CAT as the obvious quality winner and OSK as merely cheap; the better framing is that CAT has better secular duration, but OSK has better downside protection if those secular assumptions are even slightly delayed.

Key Decisions for Investors

• Pair trade: long OSK / short CAT for 3-6 months, sized modestly. Thesis is CAT’s premium multiple is more exposed to any AI-capex pause than OSK is to government-timing noise. Falsify if CAT continues to reaccelerate orders and raises margin guidance, or if OSK loses a key procurement milestone.
• Do not chase CAT above current valuation until there is evidence that data-center and power-generation demand is still accelerating. If owning, use a pullback or wait for a forward multiple reset; the risk/reward is poor if the market starts discounting peak margins.
• Treat OSK as a watchlist value/defensive hybrid, not a core defensive holding. A cleaner long only works if federal funding visibility improves and legal headlines de-escalate; otherwise the low multiple can stay low for months.
• Monitor CAT as a read-through for DE and PCAR over the next earnings cycle. A deceleration in CAT dealer inventory or backlog would be a leading indicator to trim broader industrial exposure.
• If you want express downside protection on the AI-industrial narrative, consider CAT put spreads into the next earnings date rather than outright short stock; the catalyst is event-driven, and the premium is justified only if guidance turns.