Tariff Volatility: The Inventory Decisions That Actually Matter in 2026
Most supply chain leaders report real tariff impacts this year, and KPMG’s 2026 tariff survey finds 55% of US businesses now plan further price increases in the next six months just to hold margin, per KPMG’s tariff survey coverage. The 2026 playbook has converged around a few moves: diversify sourcing (alternative suppliers have already replaced about two-thirds of the US-China trade gap, per KPMG’s 2026 trade outlook), renegotiate supplier contracts, nearshore where the math works, and carry a bigger inventory buffer, which nearly three-quarters of small and mid-size businesses are now doing by extending their planning horizons, per DHL’s mid-year 2026 pulse survey. Here is what the trend pieces skip. Buffer inventory is a loan taken out against a guess, and most companies making these moves have not priced the guess.
Why isn’t “just build buffer stock” a tariff strategy?
Pre-tariff stockpiling locks in landed cost, and it locks up working capital, warehouse space and obsolescence risk in the same motion. The arithmetic that matters: carrying cost (realistically 18 to 25% annualized once you count capital, space, shrink and obsolescence) against the tariff delta you’re avoiding, times the probability the tariff actually lands, times the months you’ll sit on the stock.
A worked example, deliberately simple. Say $1M of inventory pulled forward six months ahead of a possible 10% tariff, at a 20% annualized carrying cost:
| Scenario | Carrying cost (6 months) | Tariff avoided | Net result |
|---|---|---|---|
| Tariff lands as feared | $100,000 | $100,000 | Break-even, before obsolescence risk |
| Tariff lands at half the rate | $100,000 | $50,000 | $50,000 loss |
| Tariff delayed or dropped | $100,000 | $0 | $100,000 loss |
| Tariff lands, stock sits 9 months instead of 6 | $150,000 | $100,000 | $50,000 loss |
The hedge only pays when the tariff lands in full AND the stock turns on schedule. A 10% tariff avoided by carrying nine extra months of inventory at 20% carrying cost is a loss wearing a decisive expression.
Does nearshoring pay off?
Moving supply from Asia to Mexico cuts an ocean transit of 25 to 40 days down to a 2-to-5-day truck lane. That isn’t just speed. It’s forecast error reduction, because you are now forecasting one month out instead of three, and it’s working capital released from the water. What the celebration pieces omit: 12 to 24 months of dual-running costs, new-supplier quality curves, and the tooling and qualification bill. Nearshoring pays back on lanes with high demand volatility, high tariff exposure or high obsolescence. It’s a lane-by-lane portfolio decision, not a slogan.
How do you decide between stockpiling, nearshoring, or passing tariffs through?
Map every lane on two axes: tariff exposure and demand volatility.
Low on both means do nothing and price it through. High exposure with low volatility justifies tactical stockpiling, after the carrying-cost math above. Low exposure with high volatility is a forecasting problem, not a tariff problem. High on both is where re-sourcing and nearshoring candidates live. Do nothing and pass it through is a legitimate quadrant the consultancies undersell: if your competitors eat the same tariff, margin-neutral pricing beats a warehouse full of depreciating hedges.
The urgency behind this is real and accelerating. The share of companies in formal reshoring planning or active execution has climbed to 26%, up from just 10% six months earlier, per KPMG’s tariff survey data. That is not a slow-moving trend anyone can wait out.
Our take: inventory is a bet against your own supply chain
Whenever we talk to corporations, we share our own view of inventory. In the ideal supply chain, the just-in-time one, you carry no inventory at all, because you either have a clear view of future demand or you have infinite production capacity and zero lead times. Reality offers neither. So your inability to predict the future, multiplied by the time your supply chain needs to react, is the rough cut of the inventory you must hold. Everything above that is a choice.
And that choice has a cost beyond storage. Inventory drags profitability down. The working capital stuck in it could almost always be used better: generating demand, reducing cost, shortening lead times, raising productivity. Every one of those levers beats holding stock.
Why do tariffs work as barriers, and when do they lose their power?
A tariff is, economically, a barrier: a device governments use to stop corporations doing what they would normally do. In the current case, the United States is trying to drive internal production and employment by discouraging offshoring. The downside is the impact on consumption, because things get more expensive, and this is where price elasticity decides the outcome. Where goods can be sold at higher prices without losing volume, tariffs bite less and lose their power as policy.
Connect that back to inventory and the picture sharpens. Corporations building tariff buffers are betting against an unpredictable future. If the tariffs hold and nearshoring takes off, consumers absorb a permanent price increase and cost-push inflation follows. If the tariffs don’t stick, the companies that built inventory covers get no windfall and still pay for the working capital they borrowed, at financing costs that remain elevated. This is a classic supply chain trade-off, and the best strategy is usually the one nobody in the boardroom is happy about.
What should you instrument before you decide?
Landed cost per lane, actually maintained rather than the 2023 spreadsheet. Inventory months by origin. And the DIFOT you can hold through any transition, because service failure during a re-sourcing ramp costs you customers at the exact moment you’re least able to win them back. The boring metrics are what make the bold moves survivable, the same discipline we described in our reality check on agentic AI in supply chain: the tool or the tariff strategy is only as good as the data underneath it.
