AI Modeling Tariffs: Reshoring vs Price Hikes

A lot of manufacturers are treating tariffs like a blunt-force cost problem: eat it, raise prices, or move production. The latest ISM supply chain planning forecast makes that mindset visible in numbers. 86% of manufacturers plan to pass on at least some tariff-related cost increases, with 32% passing all of them and 42% splitting the difference (some price increases, some margin absorption). Meanwhile, only 36% are actively looking to shift production to the U.S.

That gap matters. Not because reshoring is “better,” but because it signals something procurement and supply chain teams know in their bones: reshoring is slow, complex, and full of second-order costs. When you’re staring at a tariff line item that hits next quarter, price action looks faster than network redesign.

Here’s the problem with defaulting to price hikes: you’re guessing. You’re guessing how demand will respond, which SKUs can tolerate price increases, which suppliers will break first, and whether the trade environment will shift again. In this part of our AI in Supply Chain & Procurement series, I’ll take a clear stance: if your tariff response isn’t model-driven, it’s a revenue and margin risk you can avoid. AI doesn’t remove uncertainty—but it turns it into scenarios you can act on.

What ISM’s forecast really says: “price first, redesign later”

The central message is that manufacturers see passing costs to customers as the primary lever. The ISM data shows most leaders are preparing to adjust sales prices rather than radically reconfigure supply networks.

A few specifics from the forecast are worth sitting with:

• 86% plan to pass on at least some cost increases tied to tariffs.
• 32% plan to pass along all tariff-related costs into prices.
• 42% plan a hybrid approach (some price, some margin).
• Only 6% say tariffs won’t affect their costs.
• Raw material prices averaged +5.4% in 2025, with +4.4% projected in 2026.

Those numbers align with what many of us saw through 2025: a lot of “wait and see” behavior on capital and hiring because tariff rules were shifting. With the Supreme Court reviewing legality, leaders are hoping for fewer surprises—and they’re preparing commercial actions while they wait.

My take: price hikes are not a strategy. They’re a symptom of a strategy gap.

Price is what you do when you don’t have time—or tools—to quantify alternatives.

Why reshoring isn’t the default (even when tariffs hurt)

Reshoring is less common because total landed cost is only half the story. Even if tariffs make offshore sourcing more expensive on paper, reshoring triggers a cascade of operational changes:

• New supplier qualification and audits
• Tooling moves and validation
• Labor availability and wage differences
• Different logistics patterns (inbound, warehousing, last mile)
• Compliance, ESG, and reporting changes
• Ramp risk (yield, scrap, service failures)

That’s why 64% don’t intend to bring production to the U.S. to avoid tariffs, according to the forecast. For many, it’s still cheaper to stay offshore—or diversify to a less tariff-impacted geography.

This is where AI belongs: not as a buzzword, but as a practical way to compare “price it” vs “redesign it” using the same financial and risk logic.

The hidden math: tariffs create demand volatility, not just cost

Tariffs don’t only change your costs—they change your demand curve. When manufacturers raise prices, customers don’t respond uniformly. Some SKUs are resilient; others are fragile. Some segments trade down. Some shift channels. Some delay purchases until after peak season. And because it’s mid-December 2025, procurement teams are also dealing with the annual reality: budgets reset in January, promotions get planned, and suppliers renegotiate.

Here’s what tends to go wrong when tariffs hit and pricing is the main lever:

1. One-size pricing: Raising prices “across the board” because it’s easier than SKU-level decisions.
2. Lagging signals: Waiting for sales to drop before adjusting supply.
3. Inventory whiplash: Ordering too much “just in case,” then discounting later.
4. Supplier strain: Pushing suppliers for concessions without understanding their financial resilience.

A useful one-liner I’ve found for executives is this:

If tariffs force you to guess demand, you’re paying for uncertainty twice—once in higher costs, and again in bad decisions.

Where AI forecasting actually helps (and where it doesn’t)

AI demand forecasting helps most when you have messy drivers. Tariffs create exactly that: price changes, lead-time changes, substitution, channel shifts, and customer behavior shifts.

AI can help by:

• Learning elasticity patterns by SKU/segment/channel (not perfect, but far better than averages)
• Detecting early demand shifts from orders, quotes, and customer service signals
• Stress-testing forecasts across price scenarios (5%, 8%, 12% increases)
• Translating demand risk into inventory policy changes (safety stock, reorder points)

AI doesn’t help if your inputs are garbage. If pricing changes aren’t tagged, promotions aren’t recorded, or lead times are fantasy numbers, you’ll just automate confusion.

AI-driven scenario planning: reshoring vs “China+1” vs price actions

The right way to decide isn’t “reshore or not.” It’s “which mix of actions minimizes total risk-adjusted cost.”

Most manufacturers should be running at least three scenarios for each major product family:

1. Price-pass scenario: pass 100% (or a set percentage) of tariffs into price
2. Network-shift scenario: move volume to alternate countries/suppliers (a “China+1” or “region+1” approach)
3. Reshore/nearshore scenario: shift a defined share of production domestically or nearby

The mistake is evaluating these scenarios with one spreadsheet per function. Finance looks at margin. Supply chain looks at lead time. Procurement looks at unit cost. Sales looks at quotas. Everyone “wins” locally—and the company loses globally.

What an AI-enabled tariff response model should output

A practical AI scenario model should produce decision-grade outputs, not pretty dashboards:

• Total landed cost by SKU and supplier (including duties, freight, inventory carrying cost)
• Service risk (probability of stockout, fill rate impact)
• Lead-time distribution, not a single average (P50/P90 matters)
• Supplier risk score combining financial health + geopolitical exposure + capacity risk
• Demand impact estimates under price changes (by segment)
• Working capital impact (inventory and payables implications)

If your tool can’t quantify working capital and service trade-offs, it’s not scenario planning—it’s reporting.

A concrete example (simplified, but realistic)

Say a product line has $50M annual revenue. Tariffs increase a key input cost by 8%.

• If you pass the full increase via price, you might protect gross margin—but demand could drop 3–7% depending on elasticity.
• If you absorb it, revenue holds—but margin compresses immediately.
• If you shift to an alternate country, unit cost may rise 2% (higher base cost) but avoids the tariff, while lead time might increase 10 days during ramp.
• If you reshore, base unit cost might rise 12% but lead time drops and service improves—after a 9–18 month transition.

AI adds value by calculating the expected value of each path under uncertainty—then showing what changes the decision. Sometimes one variable dominates (e.g., elasticity). Sometimes it’s lead-time variance. Sometimes it’s supplier solvency.

Procurement’s playbook: 6 moves to reduce tariff impact without panic pricing

The fastest wins are usually procurement moves paired with AI visibility. You don’t need a multi-year transformation to stop bleeding.

Here are six actions I’d prioritize for Q1 2026 planning cycles:

1. Segment your tariff exposure by “should we care?”

   • Focus on the 20% of SKUs driving 80% of duty impact.

2. Build a “duty-aware should-cost” model

   • Should-cost that ignores duties isn’t should-cost anymore.

3. Renegotiate with facts, not frustration

   • Use supplier benchmarking and clean volume forecasts to pursue shared savings.

4. Dual-source the components that create schedule fragility

   • Pick the parts that stop the line, not the parts with the loudest complaints.

5. Re-parameterize inventory policies after pricing changes

   • If price goes up, demand variance usually goes up too. Safety stock must reflect that.

6. Create a monthly tariff scenario cadence

   • One meeting, one model, one owner. Update assumptions monthly until policy stabilizes.

These are also ideal areas to deploy AI in procurement: supplier risk monitoring, demand sensing, and scenario simulation.

Why capital spending isn’t spiking (and what that signals)

ISM’s forecast points to “slight optimism” for 2026: manufacturing revenues expected to rise 4.4%, and capital expenditures expected to rise 3%. Yet the tax incentives from the One Big Beautiful Bill Act don’t appear to be driving major capex: 59% of manufacturers reported no effect on planned spending, and 20% said they’re reducing capex.

Read between the lines: leaders don’t want to commit to large moves when the rules might change again.

This is exactly why AI-driven scenario planning is a safer bridge than “big bang” reshoring decisions. You can:

• Start with supplier diversification and inventory optimization
• Quantify the benefits of regionalization before moving tooling
• Build confidence with small, reversible steps

Reshoring can be right. But “reshore because tariffs exist” is lazy thinking.

How to decide: a simple tariff response scorecard

A good decision framework balances speed, cost, and resilience. Use a scorecard that’s blunt enough to drive alignment:

• Time to impact: days/weeks vs quarters/years
• Margin protection: gross margin delta under scenarios
• Revenue risk: expected demand loss from price actions
• Service risk: fill rate and lead-time variability impact
• Execution risk: qualification, ramp, and operational disruption
• Strategic value: IP protection, regulatory risk, sustainability goals

AI helps by turning each category from opinion into probability. Not perfect probability—usable probability.

What to do next (and the question every leader should answer)

Tariffs are pushing manufacturers toward price hikes because it’s the fastest lever available. ISM’s numbers make that plain: most companies plan to pass costs, while reshoring is still a minority move. The smarter response is a portfolio approach—pricing where the market allows it, diversification where risk is concentrated, and selective reshoring where strategic value is real.

If you’re building your 2026 plan right now, don’t start by arguing about reshoring. Start by agreeing on the model that will judge it. In this series, we keep coming back to the same theme: AI in supply chain and procurement is most valuable when it turns messy trade-offs into decisions you can defend.

The forward-looking question I’d put to your team is simple: if tariffs change again in 90 days, will you be reacting—or will you already have the scenarios ready?