Talent, Not Arbitrage: Ending Low‑Wage Replacement Without Hurting High‑End Hiring

The real problem in high‑skilled immigration isn’t the researcher walking into an AI lab at a market wage. It’s the mass body‑shopping model. That model wins contracts by filing large numbers of visa petitions at the lowest allowable wage levels, stationing employees at client sites, and skating through loose oversight. Because the government’s Level 1 and Level 2 wage bands sit below the local median, the model competes most directly with trainable U.S. juniors and mid‑career workers who could be reskilled. When a client asks for “five Java devs by Monday,” the body‑shop undercuts on price, delivers immediately, and shuffles people between projects. That is the abuse surface we should close.

One story that keeps the status quo intact is the claim that Indian engineering education is categorically superior, so of course the best value is overseas. The reality is more ordinary. The very top institutions anywhere produce excellent graduates; the middle varies everywhere. Policy should not reward origin myths. It should reward verifiable skill and fair pay. If an employer will pay at or above the local median for a clearly defined job, fine. If a business model only works at sub‑median wages under weak supervision, it should face stricter rules and higher costs.

The goal is straightforward: end low‑wage replacement while preserving genuine high‑end hiring. That requires a few disciplined changes aimed at behavior—wages, placement practices, and accountability—not nationality.

Start with wages. Make the default for new, cap‑subject visas a median‑or‑better wage floor in the local labor market. Allow a narrow exception for true apprenticeships that start modestly but must step up to the median within a year. This one change removes the built‑in discount that fuels substitution. Pair it with wage‑tiered selection in the annual cap: petitions offering higher relative wages go to the front of the line and get faster processing. Employers who truly need specialty talent will signal that need in the cleanest way—by paying for it.

Then modernize rules built for another era. The “exempt” test for H‑1B‑dependent firms still uses a salary threshold set in the late 1990s. It was meant to protect against displacement; today it is too low to do that job. Replace it with a threshold tied to wage percentiles and index it annually. Keep the exemption based on pay, not credentials. A master’s degree is not a labor‑market protection.

Next, fix third‑party placements. Work performed at a client site should be treated as a higher‑risk category with higher standards and clearer accountability. Require real contracts and work orders up front, not vague letters. Enforce bench‑pay by requiring escrowed payroll so workers are not left unpaid between projects. Create joint liability so the end‑client shares responsibility for wages and compliance when they host the worker. Random site checks should be routine. The goal isn’t to ban third‑party work; it’s to make it honest and fairly priced.

Transparency will do quiet, durable work. Publish, by employer and metro, the distribution of wage levels in labor‑condition applications and match that to final approvals. Let everyone see who files most of their requests at sub‑median wages year after year. Most firms will adjust before regulators intervene.

Add a layoff guardrail that passes the straight‑face test. If a company—or the client that will host the worker—has laid off people in the same occupation and location in the past few months, new cap‑subject filings in that occupation should pause unless the firm can show it offered retraining or attempted to rehire qualified workers first. This nudges companies to invest in existing talent and treats visas as a complement to, not a workaround for, domestic hiring.

Fees should reinforce the same logic. Keep them low for high‑wage, direct‑hire roles and higher for sub‑median filings and third‑party placements. If policymakers adopt a very large new fee, carve out relief for truly specialized, top‑tier roles so we don’t punish the part of the market that isn’t the problem. Price the arbitrage, not the innovation.

To keep the innovation channel open, create a Trusted Sponsor lane. Employers that consistently pay at the top of the market and have clean records should get faster adjudication, easier portability for their workers, and a clear path to permanent residence. The message is simple: if you hire transparently, pay well, and play by the rules, we’ll get out of your way.

Now close the escape hatch: offshoring and near‑shoring. If we only raise domestic costs, some work will move to global delivery centers or just across the border. Put guardrails where public money touches private decisions. Recipients of federal subsidies, tax credits, or major IT/AI contracts should commit not to move covered functions offshore or near‑shore below specified wage tiers for a defined period, or face clawbacks. Require disclosure when customer‑facing support or essential back‑office functions are moved abroad, and restrict eligibility for new federal awards for firms that do so. For very large multinationals, tighten tax treatment of below‑median related‑party IT services that immediately follow domestic layoffs. None of this forbids global teams. It removes the one‑way cost advantage created by public dollars and tax asymmetries.

Finally, rebuild the entry ramp for U.S. talent. Use the training fees already collected from visa users to fund large‑scale, paid apprenticeships and first‑job programs in software, data, security, and AI operations. Tie a portion of visa usage to domestic hiring: for every set number of visas, hire a smaller number of U.S. juniors or apprentices in the same occupation and city, or pay an alternative levy into the training fund. This ensures employers who benefit from global mobility also invest in the next local cohort.

You will know this approach is working when three things show up in public data. First, the share of new approvals paid at or above the local median rises, and Level‑1 filings shrink to a narrow apprenticeship track. Second, third‑party placements don’t disappear but become smaller, better paid, and better documented. Third, domestic juniors actually get hired—because it becomes cheaper to train them than to work around them. That is the right equilibrium: a labor market that welcomes true specialists from anywhere while rebuilding a credible first rung for people starting their careers here.

---

I’m writing this from Seattle, where Amazon and Microsoft set the tone for the industry that surrounds us. If the goal is to end low‑wage replacement without kneecapping high‑end hiring, the two biggest players should lead—publicly, measurably, now. We don’t need to wait for Congress to act for basic fairness to become standard practice.

Here’s the pledge I’d ask them to sign and publish:

  1. Pay at or above the local median (Level 3+) for cap‑subject hires; use Level 2 only for time‑boxed apprenticeships that step to Level 3 within 12 months.

  2. Support wage‑tiered selection in policy—and mirror it internally by prioritizing higher‑wage, higher‑skill roles.

  3. Clean up third‑party placements: real contracts up front, escrowed bench‑pay, and joint liability with client teams for compliance.

  4. Layoff guardrail: no new H‑1B filings in an occupation and city within 180 days of related layoffs, unless retrain‑and‑rehire efforts are documented.

  5. Radical transparency: publish quarterly wage‑level distributions and conversion rates to permanent roles.

  6. No offshoring/near‑shoring of federally funded work or functions supported by public incentives; accept clawbacks if breached.

  7. Build the entry ramp: fund paid apprenticeships at scale and hire a fixed ratio of U.S. juniors alongside any continued visa use.

If Amazon and Microsoft codify these rules and live by them, others will follow. It won’t solve everything, but it will narrow the abuse surface, keep the innovation channel open, and—most important—give our kids a fair first step into this industry. It’s time.

---

Compete on Talent, Not Arbitrage: An Open Note to Global IT Services Firms

This isn’t a denunciation of India‑founded integrators; it’s an invitation to lead. You built a global delivery engine that can move at scale. But the part of the model that depends on sub‑median wages, client‑site placements under loose supervision, and easy offshoring has become the weakest link in public trust—and the sharpest edge cutting into the first rungs of U.S. careers. If you believe your value is expertise and execution, not labor arbitrage, then say so—and prove it.

Here’s the standard worth pledging to, openly and measurably:

  1. Compete on wages, not discounts. Pay at or above the local median (Level 3+) for cap‑subject roles. Use Level 2 only for tightly defined apprenticeships that step up to the median within 12 months.

  2. Make third‑party work accountable. Provide real contracts and work orders up front; accept random site checks; ensure bench pay is actually paid (escrow it if needed).

  3. Share responsibility with clients. Agree to joint liability for wage and compliance when your people sit on a client team.

  4. Honor a layoff guardrail. No new cap‑subject placements into an occupation and city where you—or your client—recently cut similar roles, unless retrain‑and‑rehire attempts are documented.

  5. Be radically transparent. Publish, by metro and occupation, your wage‑level distributions and conversion rates to permanent roles at clients.

  6. Train and transfer. Stand up paid U.S. apprenticeships at real scale and offer conversion rights—a clear path for contractors to become client FTEs without punitive fees.

  7. No mobility choke points. Drop non‑competes, no‑poach, and punitive liquidated‑damages clauses that trap workers. Compete by being a great place to work.

  8. Guardrails on offshoring and near‑shoring. For publicly subsidized or federally contracted work, commit to onshore delivery; accept clawbacks if you move covered functions abroad below agreed wage tiers. For private work, disclose when and why functions shift cross‑border.

  9. Invest where you hire. For every N visa placements in a metro, co‑hire M local juniors or apprentices—or pay a training levy that funds those seats.

  10. Back the rules you live by. Publicly support wage‑tiered selection, modernized protections against displacement, and tougher client‑site oversight—so good actors aren’t undercut by race‑to‑the‑bottom vendors.

Why sign? Because this is how you secure license to operate in your most important market, win preferred‑vendor status with regulated clients, attract top talent on both sides of the ocean, and de‑risk a regulatory tide that is clearly moving toward wage‑based selection and tighter third‑party controls. It’s also how you move your brand up‑market—from “people on seats” to “outcomes with accountability.”

And if you won’t sign? Many of your largest customers will start baking these terms into RFPs and framework agreements anyway. They’ll reward firms that make this shift first. Lead now and you help write the standard. Wait, and you’ll be measured against it without the reputational upside.

Seattle will hold you to this—because our kids deserve a fair first step, and because the best integrators already compete on talent, not arbitrage. If you are one of them, prove it in public.