How incentives vary by location: the value is in what you can use

Manufacturing incentives are often compared as though every grant, tax credit, loan and infrastructure promise were interchangeable. They are not. Scout Motors’ South Carolina package shows that public infrastructure can solve risks that cash cannot. Micron’s Gujarat package shows how direct strategic support can be substantial but still depend on approved spending and milestones. Lucid’s Saudi package demonstrates that financing and demand commitments can matter more than a tax holiday. The practical comparison is the Five-Gate Usability Test: identify the form of support, the approval gate, the timing, the company’s ability to use it and the obligations required to keep it.

Scout Motors picked Blythewood, South Carolina, for a $2 billion factory expected to create at least 4,000 jobs. The state appropriated $1.291 billion to support the project.

That sounds like an enormous cheque.

It was not. The package was principally land, roads, rail, water, wastewater, soil stabilisation and a publicly owned training centre. Another $200 million was structured as a repayable loan. In 2026, South Carolina legislators ordered scrutiny of a further $150 million request connected with site-preparation and road overruns, including questions about soil, wetlands and whether a flatter site had been available.

The distinction matters because a road is not cash, and neither is a water connection. Yet if a factory cannot be built without them, they can be worth more than cash. They remove the problem that would otherwise stop the project.

This is the question hidden inside almost every factory-incentive announcement: what is the support really worth to the company that has to build, hire and produce?

The usual answer is to rank locations by headline tax-credit rates, programme counts or the largest number in a press release. The C2ER State Business Incentives Database lists 2,546 active state incentive programmes in the United States alone. That includes 705 tax credits, 626 grants, 526 lending programmes and 332 exemptions.

A long menu is not a meal.

A programme count tells a manufacturer what lawmakers have created. It does not tell them whether their factory qualifies, whether the money arrives before construction costs bite, or whether the benefit survives a slow production ramp. The US International Trade Administration makes a similar warning in its 50-state manufacturing guide: it is not exhaustive and cannot replace project-specific tax and legal analysis.

What surprised us in comparing the cases was how often the apparent size of a package mattered less than its route from announcement to use. A benefit can be delayed, taxable, dependent on a milestone, limited by tax liability or lost when a factory misses its commitments.

GI Network's view: The right question is not “What has the government offered?” It is “What portion of this support survives the project timetable, tax position, approval process and downside case?”

Five gates between a headline and a factory

The best way to compare manufacturing investment incentives is the Five-Gate Usability Test.

First: what is it? A cash grant, a refundable credit, a non-refundable credit, a loan and public infrastructure do different jobs. A refundable credit can still pay value when a company has little tax due. A non-refundable credit can only reduce tax that the company actually owes.

Second: can the project qualify? The precise location, number of jobs, wage levels, investment size, sector and production output can all matter.

Third: when does it arrive? Support paid after equipment purchases, construction or hiring milestones may be valuable, but it does not fund the first bill.

Fourth: can the company use it all? A factory in its early years may not have enough corporate-tax liability to use a large credit.

Fifth: what must happen to keep it? Job, investment, operating, wage and production commitments can turn an apparent benefit into a continuing obligation.

Here is a simple comparison of the featured packages. It does not pretend to calculate a precise cash value where the evidence does not provide one. Instead, it shows where the usable value changes.

| Package | Headline support | What changes usable value | Five-Gate result |

|---|---:|---|---|

| Scout Motors, South Carolina | $1.291 billion appropriated | Support is mainly site and public infrastructure; $200 million is repayable; site conditions and overruns affect delivery | Valuable if infrastructure removes build risk, but not equivalent to cash received by Scout |

| Micron, Gujarat | India: 50% of project cost; Gujarat: another 20% | Approved project, eligible expenditure, signed milestones and disbursement arrangements | Large potential value, but only through approved spending and milestone completion |

| Lucid, Saudi Arabia | Up to $3.4 billion over 15 years | Mix of incentives, financing, workforce reimbursements, industrial-estate support and vehicle-purchase agreement | Value depends on the mix, timing and performance of each element, not the aggregate estimate |

| Ford, South Africa | 20% non-taxable cash grant for qualifying OEM assets | Application timing, qualifying assets, production, employment and B-BBEE requirements | Transparent formula, but usable only if operating requirements continue to be met |

| Intel, Magdeburg | €9.9 billion announced willingness to support | European state-aid approval, proof of financing gap and corporate project decisions | A political commitment remains conditional until the approval and financing gates are passed |

Put the cases side by side and a pattern appears that the announcements do not state plainly: incentives are not prizes handed over for choosing a location. They are contracts for a future outcome.

That is also why governments accept conditional incentives. They are not simply trying to make a company feel welcome. They want factories built, people employed, skills developed, production maintained and public infrastructure used. Conditions are the public sector’s attempt to avoid paying full price for a promise that never becomes an operating plant. The conditions can frustrate a company. They also explain why taxpayers tolerate the package in the first place.

Blythewood. Photo: Calistemon / Wikimedia Commons, CC BY-SA 4.0.

Micron’s money came with milestones

Micron, the US semiconductor company, committed $825 million to an assembly-and-test facility in Gujarat, India. The combined project cost with government partners was approximately $2.75 billion. India supplied 50% of project cost and Gujarat incentives represented another 20%.

The facility opened on 28 February 2025.

Micron’s Gujarat facility illustrates milestone-based strategic support for an approved semiconductor project. Photo: Buttonfreak / Wikimedia Commons, CC BY-SA 4.0.

This was not a general tax break for any company that happened to arrive. It was strategic, sector-specific capital support for an approved semiconductor project. Milestone and disbursement agreements were signed after approval.

That distinction is doing a lot of work. Micron’s own announcement establishes its investment commitment, while the research record also identifies the subsequent milestone and disbursement arrangements. In other words, the company announcement gives the headline; the agreements determine the route to value.

Imagine you are financing a new factory. A promise to cover part of the project cost is obviously attractive. But it changes character if payment follows approved expenditure. It may reduce the final cost of the plant. It may not fund the equipment order that has to be placed first.

The lesson is not that milestone-based support is weak. It can be exceptionally powerful. It is that a board should model it as reimbursement or phased capital support, not as unrestricted money sitting in the bank on day one.

Saudi Arabia did not offer Lucid just a tax break

Lucid, the electric-vehicle maker, received a Saudi package estimated at up to $3.4 billion over 15 years for its King Abdullah Economic City facility. It combined incentives, workforce reimbursements, industrial-estate support and financing, including up to approximately $1.4 billion from the Saudi Industrial Development Fund.

The package also included a government agreement for at least 50,000, and potentially 100,000, vehicles over ten years.

Lucid’s Saudi package combines financing, workforce support, industrial-estate assistance and a government vehicle-purchase agreement. Photo: Matti Blume / Wikimedia Commons, CC BY-SA 4.0.

The interesting part is the shape of the package. A loan helps finance construction. Workforce reimbursements can reduce the cost of building a local team. Industrial-estate support can address the practical challenge of operating at the site. A government vehicle-purchase agreement can support demand.

None of those pieces should be casually added together as if they were identical cash grants. The $3.4 billion estimate is an upper boundary across 15 years and several forms of support. The government agreement and the Saudi Industrial Development Fund financing are crucial corroborating components of the package described in the research record, but each component has its own timing and conditions.

This is where most people stop looking. They see a large incentive figure and assume tax relief is the main attraction. For a manufacturer trying to fill a factory, a credible demand commitment or concessional financing may matter more.

The county line can change the deal

Even within the United States, usability can change dramatically inside one state.

New York’s Excelsior Jobs Program offers fully refundable credits for as many as ten years. Texas JETI can limit school-property taxable value by 50% for ten years, with investment requirements ranging from $20 million to $200 million and job thresholds from 10 to 75 depending on county population.

Neither is automatically better.

A capital-heavy project with substantial taxable property may place greater value on the Texas structure. A manufacturer with limited early corporate-tax liability may prefer a fully refundable New York credit. The point is not to crown a winner before seeing the factory model.

Georgia’s 2025 schedule makes the difference even clearer. Tier 1 projects could receive $4,000 per job for five years, with excess credit usable against withholding. Tier 4 projects received $1,250 per job, limited to 50% of tax liability. For 200 qualifying jobs, that is a nominal $4 million versus $1.25 million over five years.

The higher figure matters. The ability to use excess credit against withholding may matter just as much.

For a closer look at one of the eligibility questions behind these awards, read what the but-for clause means for factory incentives.

Ford shows why a clear formula still has teeth

Ford invested R15.8 billion in its Silverton operation and supplier tooling in Pretoria, South Africa. The investment lifted capacity to 200,000 vehicles annually and added 1,200 direct jobs. Ford also benefits from the adjacent Tshwane Automotive Special Economic Zone, which provides supplier infrastructure and export logistics.

South Africa’s Automotive Investment Scheme offers a non-taxable cash grant of 20% of qualifying productive assets for original equipment manufacturers, and 25% for component and tooling companies.

That is unusually clear. It is also conditional.

OEMs generally must reach and maintain 50,000 units annually, preserve base employment and satisfy B-BBEE requirements. Applications must arrive at least 120 days before production for OEMs and 90 days before production for component or tooling manufacturers.

The percentage is the easy part. The operating plan is the hard part.

Ford’s case challenges a common assumption: that transparent, formula-based incentives are somehow uncomplicated. They are easier to understand than a negotiated political package. They still demand that the company gets the timing right and keeps delivering against production, employment and transformation requirements.

That is why what manufacturing projects are eligible for incentives is only a starting point. Eligibility is not a box ticked at announcement. It is often an operating discipline that lasts for years.

What Intel still teaches investors

Germany announced willingness to support Intel’s proposed semiconductor complex in Magdeburg with €9.9 billion. The planned complex was valued at more than €30 billion. The support was conditional on European state-aid approval and proof of a financing gap.

Intel postponed the project in September 2024.

That is enough to show the danger in treating political support as project capital before its gates are passed. European Union rules require aid to demonstrate necessity, proportionality and a financing gap. In plain English, the company must show why the support is needed, why the amount is not excessive and what shortfall it fills.

A political announcement can be meaningful. It can signal strategic intent and improve the odds of a project proceeding. It is not the same thing as an approved, paid incentive.

The evidence does not prove that conditional incentives are bad. In fact, Micron and Ford show why they can be central to a credible industrial plan. The lesson is narrower and more useful: uncertainty should be priced before the company counts the support.

What operators should do before they choose

Build an incentive register before making an irreversible site decision. For every state, local, utility and sector programme, record eligible expenditure, job and wage obligations, application deadline, approval point, tax treatment, payment date and consequences of non-performance.

Then run each item through the Five Gates. Separate support into four lines: announced, agreed, approved and received. A lender and an equity investor may assign a different value to each line. Your internal model should too.

Treat infrastructure honestly. It is not cash. But roads, water, rail and site preparation may solve the one physical problem that turns a promising location into a buildable one.

Finally, test the workforce promise. A credit tied to jobs is only as dependable as the hiring plan. Underwriting workforce risk before talent constrains capex matters especially where employment commitments sit at the centre of the award.

What investors should ask before counting it

First-time founders often repeat the press-release total because it is visible and flattering. Experienced investors pull it apart.

They ask: Is this cash, credit, financing or infrastructure? Is the credit refundable? Does the company have the tax liability to use it? Is it taxable? Does payment follow spending? What happens if production ramps slowly? Which commitments survive a change in ownership, factory scale or timetable?

GI Network would turn a proposed incentive package into an underwriting schedule: matching every element against capex, equipment orders, construction milestones, workforce commitments and financing needs; testing the effect of tax and payment timing; separating government commitments from company announcements; and preparing the investment-committee case around support that is genuinely usable.

The boardroom tool is simple: Pass the Five Gates before you count the money. Ask what the incentive is, whether the factory qualifies, when it pays, whether the company can use it and what it must do to keep it. If one gate is weak, the headline is not yet a financing plan.