Five Weak Arguments for a US Manufacturing Policy, and Two Real Ones
The strongest case for U.S. manufacturing policy is not jobs or economic multipliers. It’s the trade deficit and China’s techno-economic challenge.
At ITIF, I have advocated for a robust national manufacturing strategy for over two decades. Let me get that on the table so that when I say most of the arguments offered for why U.S. policy should support manufacturing are not very convincing, you won’t accuse me of being aligned with Adam Posen of the Peterson Institute. Let’s not get carried away!
Weak Arguments
Before I provide two real reasons why manufacturing matters, let me rebut five common but ultimately weak arguments.
“Manufacturing delivers good jobs.”
In reality, there are not that many jobs in manufacturing. U.S. manufacturing accounts for just 7.5 percent of total jobs. From 2016 to 2025, manufacturing job creation was just 2 percent of total net new jobs. Even if the United States eliminated the manufacturing trade deficit, it would maybe boost manufacturing to 10 or 11 percent of total jobs.
“Manufacturing jobs are good jobs.”
Are manufacturing jobs good jobs? They used to be. In the early 2000s, there was a wage premium of 3 percent for manufacturing production jobs. But according to the Federal Reserve Bank, that premium has disappeared, with wages for manufacturing production workers 5 percent below those of the rest of the private sector.
To be fair, the wage premium is more significant in rural areas, as is manufacturing’s share of total employment, suggesting that one reason to support more manufacturing is as a rural development tool. But given that rural manufacturing jobs declined between 2019 and 2023, it’s not clear how much a revived manufacturing sector would do for rural economies.
“Manufacturing is critical to productivity.”
Sure, manufacturing has, on average, higher value-added per worker than the rest of the U.S. economy. But if, by some miracle, America could boost manufacturing productivity twice as fast as economy-wide productivity (assuming 2 percent productivity growth for the economy and 4 percent for manufacturing), this would boost overall productivity growth by just 2.3 percent after 10 years. And even that estimate is extremely optimistic, given that manufacturing productivity actually fell from 2011 to 2019.
“Manufacturing is critical to innovation.”
It is. Even with the vast increase in R&D in America’s information sector (think Google, Amazon, Meta, Microsoft, etc.), the majority of U.S. R&D still occurs in manufacturing. R&D is critical because it drives innovation (e.g., the introduction of new goods and services into the market), which, by definition, improves our lives. Moreover, much has been written, correctly in my view, about the importance of the tight geographical linkage in many sectors between manufacturing, R&D, and innovation.
But here’s the rebuttal: Why can’t the United States just buy the innovative products and services it needs from other nations? If Germany produces better robots or Taiwan makes better computer chips, America can just buy them. As long as global innovation is robust, the United States is fine.
“Manufacturing has one of the highest economic multipliers of any U.S. industry.”
Bureau of Economic Analysis input-output tables show roughly $1.40 to $1.90 in additional output per dollar of manufacturing output, versus much lower multipliers for most other sectors of the American economy. Having said that, unless the economy is in a deep recession and policymakers want to expand demand, who cares? If the economy is close to full employment, which it usually is, multipliers are largely irrelevant.
Actual Reasons
Okay, that gets us to the two real reasons why manufacturing is important.
The trade deficit.
Manufacturing accounts for roughly 60 percent of U.S. goods-and-services exports despite being roughly 11 percent of GDP. Without manufacturing, the trade deficit would skyrocket even more and, in contrast to what the Adam Posens of the world might have us believe, that is not a free lunch. A large trade deficit makes Americans better off in the short run because U.S. consumption exceeds production. But unless other nations are willing to subsidize American overconsumption forever, this cannot continue indefinitely.
At some point, the value of the dollar will have to fall significantly so that Americans consume less than they produce and export the difference, thereby lowering real living standards because the United States will need to import less, consume less, and export more. Putting off that day of reckoning will only make things worse. The longer America delays that adjustment, the more debt and economic burden it passes on to future generations.
Which leads to… drum roll, please… the second and BIG reason…
Without major structural change in U.S. policy, akin to the strategy America implemented at the beginning of the Cold War, U.S. national economic power industries, most of them manufacturing-based, will weaken significantly or die. In turn, this will diminish the U.S. defense industrial base and give China techno-economic leverage over America across a broad range of sectors, not just rare earth minerals.
The United States simply cannot afford to live in a world where it is dependent on China for jets, drugs, chips, machine tools, autos, and more. Dependence of that kind would give Beijing enormous leverage and hollow out the U.S. defense industrial base.
That means America needs a manufacturing competitiveness strategy. Not as social policy to create jobs. Not as environmental policy to create “green” products. Not as growth policy to boost productivity. But as national techno-economic security policy.
Rob does important work clearing away the weak arguments, but I think the two "real" reasons he lands on are still downstream of the actual structural case.
The strongest argument for U.S. manufacturing policy isn't the trade deficit or even China as a geopolitical adversary, it's that the economics of global supply chains have fundamentally broken down, and the private sector has already figured this out even if the policy world hasn't caught up.
Three Forces Making the 2000s Offshoring Logic Obsolete:
1. China's demographic cliff is not a risk, it's arithmetic.
The one-child policy's consequences are arriving now. Chinese manufacturing wages have risen roughly 5x since 2005, and the working-age population is entering structural decline as their largest generation hits mass retirement. The dependency ratio will force wages higher indefinitely. This isn't a policy problem to manage, it's a mathematical certainty that eliminates the labor arbitrage that made offshoring rational in the first place.
But the deeper point goes further than China specifically. The entire offshoring cascade, from Japan to South Korea to Taiwan to China to Vietnam, was only possible because it could repeatedly find the next large pool of impoverished, accessible labor. That cascade is now running out of runway. The remaining low-wage populations lack the infrastructure, political stability, logistics networks, and industrial base to absorb complex manufacturing at scale. Bangladesh's garment wages have more than doubled since 2010. Vietnam's manufacturing wages have tripled since 2012. There is no next China. The arbitrage that drove 40 years of globalization consumed itself in the process. The workers got richer. The wages converged. The model is spent.
2. The shale revolution changed the global cost equation permanently.
This doesn't appear anywhere in the article, but it's arguably the single biggest structural tailwind for domestic manufacturing. U.S. industrial natural gas prices are now roughly 15 times cheaper than Europe and Japan, and approximately 4 times cheaper than China. This matters because energy represents 40% of the cost of heavy manufacturing, more than labor. Germany is the cautionary tale sitting in plain sight: an industrial powerhouse that deliberately dismantled its energy base and is now paying the price with stunted industrial output. The United States made the opposite choice and most analysts haven't internalized what that actually means for manufacturing competitiveness.
Between 2014 and 2019 alone, U.S. oil production increased by 5.3 million barrels per day, effectively adding the equivalent of one Canada to the market. The shale revolution saves U.S. consumers and industry an estimated $203 billion annually. When you layer renewable energy potential on top of that, the Great Plains wind corridor, the Sunbelt solar capacity, the hydroelectric resources of the Pacific Northwest, the energy architecture of North America is moving from sufficiency to abundance in a way that has no historical precedent. Automation and greentech compound this further. When you eliminate labor arbitrage through robotics and have the cheapest industrial energy in the developed world, the math on domestic production looks completely different than it did in 2005.
3. Corporate boards have repriced supply chain risk after repeated existential shocks.
COVID didn't just disrupt supply chains, it demonstrated that single-country concentration of production could halt entire industries for years. The Strait of Hormuz, Red Sea disruptions, and Taiwan anxiety have compounded this. The question in boardrooms now isn't "is domestic production cheaper?" It's "can we survive another shock with our supply chain in the other hemisphere?" That's a different calculation entirely, and it's driving capital allocation decisions that no policy instrument could have engineered.
The old cost-benefit calculation assumed cheap transport, geopolitical stability, and predictable transit. Every one of those assumptions was stress-tested between 2020 and 2024 and failed. When CFOs now run Total Landed Cost models, incorporating shipping costs, tariff risk, inventory carrying costs, quality control, and supply chain disruption insurance, the overseas "savings" often weren't what they appeared. The financial repricing of supply chain risk is permanent because it's now embedded in corporate governance frameworks, not just CFO memory.
Manufacturing is a way to commercialize proprietary knowledge and ideas as product features. Success stories are TSMC, Largan, and many more. It's worth noting that although iPhone design is Apple's idea, component supplier Largan's gross margin exceeds Apple's by a factor of 2. Is it a weak or strong reason? Here is Largan lesson: https://www.the-waves.org/2022/02/19/high-profit-ideas-of-lens-making-a-lesson-for-startups/