By 
The division of labor depends upon the extent of the market, but the extent of the market also depends upon the division of labor. The division of labor is limited not merely by the size of the population but by the size of the market, and the size of the market is itself a function of the division of labor. The process is cumulative and self-reinforcing.
–Allyn Young, “Increasing Returns and Economic Progress” 1928
Imagine you’re a T-shirt producer. You make clothing for your town’s local boutique and work from a garage. You have a basic sewing machine. You cut fabric by hand. You might even use your kitchen scale to weigh packages. It’s labor-intensive, but manageable. And you can stop at any time.
But now imagine you’re making T-shirts for a global retailer. Tens of thousands of units. Sold in dozens of countries. That garage and home sewing machine no longer cut it.
You lease a warehouse. You install industrial cutters and multi-head embroidery equipment. You set up automated folding and tagging systems. You purchase the most advanced design software and hire a small team of creatives. You don’t box shirts by hand — you contract with a fulfillment center.
Suddenly, you’re no longer just “making T-shirts.” You’ve become a node in a vast, global production and exchange network. Your operation — its warehouse, machinery, and specialized tools — only makes economic sense at scale. It is predicated on the expectation that you will serve thousands of customers far beyond your local area. This kind of large-scale, specialized production lowers, rather than raises, the cost of each shirt. In the words of Adam Smith, “the division of labor is limited by the extent of the market.” The more shirts you produce, the more you — and your trading partners — can specialize in specific tasks. In turn, this specialization drives down average production costs. As the market expands, deeper specialization yields further cost reductions. Economists call this dynamic “increasing returns to scale” — as output grows, per-unit costs decline. Not just for you, but for everyone connected to this network, including the end buyer of t-shirts.
But here’s the kicker: the entire system rests on entrepreneurial expectations — that tomorrow will resemble today. That you will remain tapped into the network of exchange. That sales volumes will continue long enough to pay off your specialized capital investments. Your factory lease, your custom machinery and software, your supplier relationships, your shipping contracts — they all depend on the expectation that you will have access to global markets such that you can keep making and selling shirts at scale.
Now imagine that some politician in Washington declares — for one reason or another — “we’re shutting down trade.” Tariffs spike. Quotas tighten. Overnight, you find yourself severed from the global network that your business was built to serve. Suppliers withdraw. Buyers cancel. Contracts unravel as partners back away from the new, unpredictably high prices. And there’s no going back to the garage — it’s far too late for that. Your business isn’t a side hustle; it’s a full-scale operation, purpose-built for high-volume production, distribution, and exchange… not handcrafted batches sold at local farmer’s markets.
Running specialized machines for a handful of local buyers wrecks your margins. Per-unit costs skyrocket. The sunk costs remain, but the global network that justified them is gone. You’re left with infrastructure built for a world of massive integration that no longer exists. It’s like owning a tractor when all you need is a shovel, or running a commercial kitchen to pour a bowl of cereal.
So, now you face a choice.
You can’t sell enough T-shirts to cover the cost of your existing capital. But those buildings and machines can’t easily be repurposed, either. They are highly specialized; built for a specific kind of production at a specific scale.
What to do?
You could try to sell the equipment — but who would buy? Firms that once might have bought from you to expand their operations are facing the same retreat. Their market has contracted, too. You could let the machines sit, hoping the policy reverses and global trade recovers. But every day they gather dust and bleed losses. Payments don’t simply halt. The creditors will eventually come knocking.
You might repurpose the warehouse, auction off assets, or scale down and rebuild — but these are costly fallback strategies, not recovery plans. No matter what, you lose.
And this loss is not confined to just your business. The whole world shares in this loss of productive potential. Those machines, in the right context, could have created real economic value for countless others: meaningful jobs, lower prices, greater variety, etc. But this utility wasn’t intrinsic to the gears and needles — it was contingent upon their place within the global division of labor. The operation holds value only so long as it remains situated within a vast network of exchange.
Across industries, firms of all kinds face the same dilemma. From logistics hubs to component suppliers, from small manufacturers to global firms, businesses built for scale are finding that their plans no longer cohere in this new world where the scope of exchange has abruptly contracted.
This is why trade commands so much economic attention. It’s not simply the movement of goods across borders; it’s a system of interdependent plans, capital commitments, and coordinated expectations — each contingent on the health of the network as a whole.
When trade expands, it enables specialization and scale. Firms make investments, accordingly, building and acquiring capital that only make sense amidst a large network.
But when the extent of the market contracts, the system doesn’t simply take a brief pause. It begins a process of self-reconfiguration. Its components adjust. Like a muscle deprived of use, where tissues and cells begin to atrophy. The body physically changes and loses strength. When movement resumes, recovery takes time — and the muscle is seldom as strong as it once was.
Like our bodies, economic atrophy is not ephemeral. Nor is it easily reversed. Entrepreneurs adapt to a contracting trade network by investing in capital better suited to smaller-scale production. If disruption seems temporary, they might hold, picking up where they left off once trade networks re-open. But any prolonged uncertainty forces adaptation. New capital investments — fitted to narrower markets — anchor firms to a reduced capacity of output, as such investments must then be amortized over some length of time. Reinvestment in large-scale capital lags behind policy change, especially when that policy is uncertain. This lag is most damaging in long-horizon ventures (like aged wine or offshore drilling) where economic viability hinges on stable, long-run expectations.
As scale retreats, efficiency falls. Capital that was once finely-tuned to global integration is retired, repurposed, or scrapped. Thus, if and when trade openness is restored, the system may no longer produce as cheaply — or as abundantly — as it had prior to the restrictive policies.
The damage done by Trump’s “Liberation Day” tariffs, then, is not some transient dip in output or a blip in prices. It is a structural downgrade in our capacity to create. Trade is not a tap to be turned on and off, a light switch we can flip at will. It is a path-dependent architecture, built slowly over time by enabling entrepreneurs to integrate themselves into the global economic ecosystem. Once dismantled, the cost-reducing forces that global trade enabled — investment, specialization, and the accumulation of knowledge — do not simply resume from where they left off. They must be rebuilt from a diminished base, slowly and often at considerable cost to those who once benefited from, and flourished within, the extent of the market.