Yesterday we wrote about this fascinating WSJ story by Heather Somerville about the trials facing Defense Tech startups in sourcing components from outside China. "China is the dominant supplier of batteries, motors, sensors, rare-earth materials and other key components needed by U.S. defense companies," the story correctly notes, and highlights that defense startups, in particular, have a hard time sourcing components.

We have looked around for solutions to this problem, and they all fall into familiar buckets: more tax credits, more subsidies, low interest loans, and streamlining regulations. Those may all be good ideas, and should be pursued, but they are vague and don't address the specific problems of defense startups needing critical components made in the US.

Therefore we think we should add our own ideas. In doing so we tried to come up with actionable ideas that would sidestep the problems associated with getting complicated reforms passed by Congress and then operationalized by a lethargic bureaucracy.

Here they are.

The Defense Innovation Purchasing Cooperative

The cooperative is a highly underrated corporate form. Cooperatives are corporations that are owned by their members, each member having an equal share, that engage in some business for the benefit of its members. The obvious drawback relative to the standard corporation is access to financing, since they cannot sell equity to outside investors, which is why cooperatives have been mostly successful in customer-financed industries like finance and retail.

The US prides itself, rightly, on how many of its banks are credit unions: a credit union is a banking cooperative. If you or someone in your family is a veteran, you probably bank at USAA because you know how outstanding their customer service is, and you probably understand that there's a connection between care for the customer and the fact that the customer is literally a part-owner of the bank.

Cooperatives are also very useful for ecosystems that contain many small businesses to pool risk—and also modernize. During the first Agricultural Revolution of the 18th and 19th century, France maintained its status as an agricultural superpower by encouraging farmers to form cooperatives to invest in capital goods (machines and fertilizers). The descendant of these cooperatives, Crédit Agricole, is not only still around, but is the 9th largest bank in the world by balance sheet size.

The obvious benefit of the cooperative model for our purposes here, in particular the current form of the US defense procurement ecosystem, which has a handful of very large companies and a burgeoning startup ecosystem, is that they don't benefit the big guy over the small guy. This is not always a good thing, but it is definitely a good thing in the present circumstances, where the American economy and national interest has an interest in fostering a flourishing defense tech startup ecosystem.

The job of the cooperative would be to create a market for US-sourced components. By definition, the cooperative would prioritize whatever components a majority of its members would prioritize, which seems smarter than having someone at the Pentagon draw up a list.

The entry cost into the cooperative should be low for defense startups. The cooperative would raise the rest of the capital by issuing specially-structured bonds whose payouts would be different depending on the cooperative achieving certain revenue and purchasing milestones. Because the cooperative would only have to pay back the bonds once it hits certain financial milestones, there wouldn't be any risk of debt costs strangling it in the crib. The Pentagon should cajole Defense Tech investors into buying these bonds, which we conjecture wouldn't be difficult, not just because Defense Tech investors have a corporate interest in the success of such a cooperative, but because these bonds would, financially, be a hybrid security with equity-like returns, which private equity investors can find attractive and could justify to their LPs (LPs is the name for investors into private equity funds) as providing a good risk-adjusted returns. This investment would not be an act of charity, therefore.

An even more effective way to help generate those bonds would be for the government to commit itself to matching private purchases. Another way the government could help with startup capital for the cooperative, obviously, would be through direct grants.

To summarize, the cooperative would (1) have significant capital at its disposal to create a market for the kinds of components that Defense Tech startups need; (2) have aligned market incentives and lack of bureaucratic overhead to be able to source, purchase, and resell the right kinds of components. The government's role would be as a kind of midwife, facilitator and, perhaps, providing key startup capital for the enterprise.

The National Reshoring Purchasing Corporation

A common proposal to help reshoring is purchasing commitments from the Pentagon. The problem is that the Pentagon doesn't need batteries or chips or rare-earth materials, it needs drones and ships and tanks.

Hence our second idea: the National Reshoring Purchasing Corporation. The basic idea is this: the government would set up a private corporation, 100% owned by the government, which would commit to purchasing large amounts of components at above-market rates and then sell them back to participating startups at cost.

One of the key problems with reshoring is that it requires a large upfront investment. In the private market, such large capital investments can only be justified if there is a strong enough belief that there will be a market for the goods produced a long time into the future. In the present situation, only the US government has the financial firepower and lack of market incentives to be able to make this kind of commitment. This therefore justifies spending taxpayer money.

However, if we want to encourage the capital investment that is needed to produce these key parts in the US, the government must make a commitment that is not just credible on Day 1 but several years into the future. This is why we believe that it is important to create a private entity separate from the government rather than a Pentagon program.

It's also very important that this new company doesn't become a market-distorting boondoggle. Therefore, its charter should have key rules laid out in a very simple and clear way. Namely:

• Open access and standardized sales. The whole point of the NRPC is to create the kind of economies of scale that justify investing in factories to make these products. Therefore, the product lines it purchases should all be clearly standardized, and it should sell them at a single cost to any qualified customer, like a supermarket. (Qualified customer should be defined very generously: for example, any US-based company with revenue under $500 million that has had at least one defense contract in the past 5 years.) No 5 year gajillion cost-plus contract to build some obscure component that only one company can use. If defense tech startups really need such specialized components, they can find some other way to source them. The goal of the NRPC is not to fulfill all startups' procurement needs, but to create a market for off-the-shelf components. Following the 80/20 rule, the NRPC won't sell everything its customers need, nor should it, but it will sell a lot of off-the-shelf components that they can use.

• Arbitration. A big part of why defense contracting is so inefficient is that right after Bidder A wins a contract, Bidder B sues. The NRPC's charter should state that all claims by its suppliers or customers will be submitted to third-party arbitration and that by agreeing to sell or buy from NRPC they waive their right to appeal.

• Sunset clause. This is the most important one. Buying widgets at above-market rates to sell them at below-market rates is, by definition, a doomed business model. It takes all of 30 seconds for anybody with a brain to realize how the NRPC could become some unkillable boondoggle that distorts the market, rather than fosters it. Therefore, it should have an ironclad sunset clause specifying that after, say, 8 years in operation, it will shut down. The purpose of the NRPC is to create a market that can eventually be sustainable on its own after the upfront capital costs have been amortized. After its 8 years in operation, the NRPC will either have succeeded in doing so, in which case it will no longer be needed, or it will have failed, in which case it will be time to try a new approach. The charter should also provide that after shutdown, NRPC employees cannot work for NRPC suppliers or customers for two years, and provide for an extremely generous severance package upon shutdown, to make sure all the incentives are aligned.

The Regulatory Side

This is all good, but there is still important work to be done on the regulatory side for it to work.

The first and obvious one is to ban Chinese-made components. This is not practical today, so there should be a grace period. Combined with the creation of the purchasing cooperative and the NRPC, this should only increase incentives to create the supplier ecosystem we want to see.

The second, equally obvious, is permitting reform. Particularly when it comes to rare earth mining, but also generally building new factories, US regulations are onerous.

The final is not regulatory as such, but has to do with workforce development and particularly encouraging vocational and apprenticeship programs, and veteran transition programs. A simple rule change would be to make all the benefits available to college students and graduates available to those who choose to do a vocational program or apprenticeship as part of the reshoring initiative.

Policy News You Need To Know

#Labor — You should read our analysis of, and proposed solutions to, to the ILA strike.

#Immigration — The Center for Immigration Studies, which for our money produces the highest-quality work on immigration, has had an outstanding idea: the National Environmental Policy Act requires the Federal government to produce environmental impact statements for any action that significantly affects the environment. Whether or not that's a wise law, it's the law. Except that DHS has not produced an environmental impact statement on its decision to stop enforcing the Southern border. Therefore, CIS and others have sued DHS to force it to produce this report. They won. Funny, and very clever.

#Energy #Solar — Bloomberg has a fascinating exposé on how the US, once the leader in solar panel manufacturing, lost the race to China, which now dominates. Here's the article's thesis, according to its author, David Fickling: "There’s a ready explanation used by trade warriors as justification for tariffs and other bans: Beijing set out to dominate this industry, and may want to use solar energy as a weapon the way Moscow uses gas. […] But I don't think it makes a lot of sense. China has offered a lot of support to state-owned companies and the real estate sector over the past decade - but that's not been the case with solar power. In fact the industry is a bit of a basket case. […] I think the far better argument staring us in the face: solar manufacturing is fundamentally low-margin and capital-intensive, so it's really, really hard to make money. Western solar companies saw this years ago, and it's why they never invested enough." As we said, fascinating.

#Energy — Utilities often face a fairly obvious problem: consumers pay a flat rate for electricity, but the cost they pay for that electricity fluctuates, sometimes hour-by-hour. Enter dynamic pricing: charge more at peak times, charge less during down times. The idea is elegant in theory, but does it actually work? A new NBER paper looks at this and comes to two conclusions which should strike us as common sense, but are still worth seeing empirically validated and keeping in mind: (a) dynamic pricing typically works; (b) unless you make it too complicated, in which case customers no longer understand it and it doesn't accomplish its goals.

#Fertility — At the Mises Institute, Ryan McMaken has a good post noting that yes, car seat laws reduce the birth rate.

#ForeignInfluence — R Street's Adam Thierer has an op-ed on a worrying development: the EU has set up an office in San Francisco whose ostensible purpose is to lobby (or bully?) US tech companies into applying EU regulations worldwide, and not just in the EU. The EU's new "ambassador to Silicon Valley" (sic.) is quoted at saying that "the name of the game is compliance." Really? And here we thought the name of the game was innovation and value creation.

#Veterans — RAND has an interesting new report on supporting America's military and veteran caregivers, whom it calls "hidden heroes." Surely a worthy goal.

#Budget — Brookings has an interesting new paper out, finding that, historically, Congress responded directly to projected budget deficit increases by cutting spending and raising taxes—but has stopped doing so from 2004 to 2023.

#VotingRights — Governor Gavin Newsom signed a bill into law that bars local governments from enacting laws to require residents to show a valid form of identification in order to vote, Democracy Docket reports. We really try to refrain from engaging in conspiracy theorizing, but we really struggle to come up with any rationale for such a law except to facilitate and encourage cheating. Obviously in California it doesn't really matter for the Federal election, but it does to maintaining California as a one-party state.

#Milestones — Happy 100th birthday to President Jimmy Carter! He is the first President ever to become a centennarian.