80% of Firms Used Generators. The Zone Had Tax Breaks.

Angola, Peru and South Africa show why special economic zone status and generous incentives do not automatically create export industries. Ghana and newer ASEAN models suggest that durable zones are operating ecosystems, not discounted parcels of land.

In Angola, the promise was easy to understand. A Special Economic Zone, known as the ZEE, offered tax exemptions and customs facilitation. For a manufacturer deciding where to put a plant, that is the sort of offer that makes a board sit up.

Then the factory had to operate.

Around 2019, firms in Angola’s ZEE faced frequent power outages. According to research on the zone, 80% of firms relied on generators. Customs clearance also remained slow and bureaucratic. So the investor was handed relief from one cost while absorbing uncertainty in two that arrive every day: making goods and moving them.

That is a bad trade.

The story raises a deceptively simple question. If tax is lower inside a zone, why does the zone not automatically fill with factories, jobs and export orders?

The usual answer is that the offer was not generous enough. More years of exemption. Cheaper land. A smoother permit. That instinct is understandable. A tax break is visible, announceable and simple to price.

A dependable production ecosystem is harder. It is a power line that works on an ordinary Tuesday. It is a shipment that clears when the manufacturer planned. It is workers who can reach the site, learn the job and build a life nearby. It is rules that survive a change of administration.

The myth is that an industrial zone is mainly a property development with a favourable tax code. The evidence says it is closer to an operating system. If its basic functions fail, the attractive headline cannot rescue it.

The offer that cannot run a machine

Manufacturing is unusually unforgiving of interruption. A retailer may endure a slow approval process once. A factory has to buy inputs, run equipment, manage workers, clear goods and ship orders repeatedly. Every disruption becomes part of the cost of the product.

This is why the tax-holiday argument is often back to front. Tax is paid on profit. But an unreliable power supply, delayed customs clearance or a missed shipment can destroy the profit before tax is even relevant.

The International Monetary Fund put the point bluntly in an August 2026 note: special economic zone success depends far more on infrastructure, governance, regulatory efficiency and integration with the domestic economy than on tax breaks alone. That is not an argument against incentives. It is an argument about sequence.

First make production possible. Then make it competitive. Only then ask whether an incentive tips a close location decision.

What surprised us was how neatly this distinction appears across places that are often discussed separately. Angola had exemptions and facilitated customs procedures, but not dependable electricity or fast clearance. Peru offered exceptionally generous treatment, including an indefinite corporate-income-tax exemption, yet struggled to draw new foreign investment. South Africa formally designated zones, but firms inside them showed weaker export and productivity results than peers outside.

Different countries. Same missing machinery.

For an operator, this should change the due-diligence questions. Do not stop at “what tax rate applies?” Ask whether port infrastructure, roads and rail can support the intended volumes. Ask who controls electricity and water reliability. Ask whether customs throughput is predictable, not merely nominally facilitated. Ask whether the labour pool exists, and whether employees and their families can realistically live nearby, with schools where international staff are needed.

Those questions may sound less glamorous than a five-page incentive schedule. They are the questions that determine whether the schedule is worth anything.

Peru’s forever exemption

Peru offers the cleanest rebuttal to the belief that a bigger tax break must do the job.

Its early special economic zones offered generous treatment, including an indefinite exemption from corporate income tax. “Indefinite” is a powerful word in an investment presentation. It sounds like certainty. It sounds like an edge competitors cannot match.

Yet the OECD found that Peru’s early zones attracted little new foreign investment because of poor infrastructure, remote locations, constitutional hurdles and shortages of skills. The tax advantage did not disappear. It simply could not overcome the rest of the journey.

Imagine you make goods for export. A permanent tax exemption may save money after the factory works, sells and earns a profit. But remoteness affects every incoming component, every container and every recruitment decision. A shortage of skills affects output from the first shift. Poor external connections are not a line item. They are the terrain.

Zofratacna, Peru’s tax-reduced retail-goods zone, exposes a second problem. The OECD found that it mostly diverted consumption rather than creating additional activity, while risking unfair competition. A zone can look busy and still add very little to the economy. Goods move. Shops trade. Tax concessions are used. But the activity may merely have shifted from elsewhere.

Zofratacna illustrates the danger of activity that shifts consumption without creating additional export value. Photo: Daniel Reynaga / Pexels, Pexels licence (free commercial use).

Here is what the numbers do not tell you when a zone authority announces registrations or occupancy. They do not show whether a new factory would have arrived anyway. They do not show whether exports are genuinely additional. And they do not show whether the fiscal cost of the incentive is funding a lasting productive capability or a rearrangement of local spending.

That distinction matters because governments are not simply giving away a theoretical benefit. They are giving up revenue while often also funding land, access and administration. A zone should therefore be judged on additional exports, durable employment, investor survival and domestic procurement, not on the ceremonial count of businesses registered.

For a related lesson in how official support can fail to turn into usable operating cash, see why lenders discount the subsidies governments celebrate.

A label is not an ecosystem

South Africa’s special economic zones supply the next uncomfortable finding. A 2026 UNU-WIDER study found that firms in the zones had significantly lower exports and productivity than non-SEZ firms. Other measures, including assets and employment, were weak or insignificant.

This does not mean a zone caused every weak result. The evidence does not prove that. It does mean that designation itself is no performance signal.

That should puncture a common assumption in industrial zone investment. A map boundary, a zone authority and a list of privileges are administrative facts. They are not proof that an industrial ecosystem exists within the fence.

The study points towards the missing work: targeted industrial-policy enforcement and infrastructure that integrates the zone with the wider economy. Put plainly, somebody has to ensure that the zone’s promise reaches the factory floor and the export corridor. Otherwise a special status can become a sophisticated signpost pointing to ordinary constraints.

There is an even wider lesson here. Why not make the entire country one giant special economic zone? Because a zone is supposed to solve concentrated failures: a missing logistics connection, fragmented approvals, unavailable industrial utilities or a need to coordinate suppliers and workers around a production cluster. If the whole country receives the privilege without fixing those failures, the country has not become competitive. It has merely extended the discount.

A small urban zone can work, too, but only where its location solves a real coordination problem. A few city blocks are not inherently too small. They are too small if the business needs a port corridor, heavy power supply, large workforces or land-intensive production that the city cannot provide.

Ghana got the order right

The counterexample matters. If the conclusion were simply that incentives never work, it would be wrong.

A study of 328 firms between 2018 and 2021 found that Ghanaian firms registered within SEZ enclaves had higher productivity, revenue and profit than firms outside them. Ghana’s zones combined tax incentives with tangible benefits, including utilities and tariff support.

That last phrase is doing the real work. The policy was not only a promise about tax paid later. It offered usable conditions for operating now.

Ghana does not prove that every zone with infrastructure will succeed, nor does it isolate every cause of stronger performance. But it gives the argument its necessary boundary: incentives can contribute when they sit on top of functional infrastructure. They are a lever, not a foundation.

Put the Ghana, Angola and Peru cases side by side and a pattern appears that none of the reports states quite this way. Governments often treat incentives as the opening bid to investors. Manufacturers experience them as the final adjustment to a much larger calculation.

The calculation is brutally practical. Can I run? Can I hire? Can I clear? Can I ship? Can I trust the rules? Can I source locally over time? If the answer is no, lower tax may make a fragile project cheaper, but it does not make it sound.

The zones trying to make exit irrational

The newer competition in ASEAN makes this clearer still.

GIFT City in India and the Johor-Singapore SEZ, the cross-border project linking Malaysia and Singapore, are competing on institutional depth: energy reliability, financial infrastructure, legal certainty and corridor integration. Penaga Research describes the result as “sovereign anchorage”. The phrase is useful because it captures the ambition. The goal is not merely to persuade an investor to arrive. It is to make leaving operationally unattractive.

GIFT City represents the newer contest: building reliable institutions and connections that make long-term investment stick. Photo: Vijay Parmar / Wikimedia Commons, CC BY-SA 3.0.

Tax competition is easy for another jurisdiction to copy. A functioning ecosystem is not. Legal certainty takes time to establish. Skilled workers accumulate. Supplier relationships deepen. Transport links become embedded in daily decisions. Financial and trade systems learn how to serve particular industries.

That is also the historical lesson from China, Korea, Malaysia and Singapore. Their zones served as testing grounds supported by infrastructure, industrial clustering and supply linkages. China’s zones improved productivity, employment and investment through these clustering effects, not through tax generosity alone.

The interesting part is that this changes how “investor selection” should be understood. A zone should not only chase the largest name or the quickest land sale. It should seek firms that can buy from local suppliers, train workers, use the corridor repeatedly and attract compatible businesses. A footloose project may fill a plot. An anchored manufacturer helps build a system.

When the tax break really does matter

There is a fair objection. If two locations are equally reliable, connected, well governed and well staffed, tax can decide the contest.

Of course it can. It can improve returns, reduce early risk and influence the marginal choice between otherwise credible sites. Ghana’s evidence is consistent with precisely that role.

But that is a narrower claim than the one too often made. It is not “tax breaks build successful zones.” It is “tax breaks can help a functioning zone win.” The difference is the entire investment case.

A manufacturer should also test trade access before treating any zone as an export platform. The country’s trade agreements, its port and corridor capacity, and the practical ease of getting goods through customs determine what market the factory can actually reach. An export order that cannot be collected or cleared on time is not much of an advantage. The best export order can still leave you cash-poor for much the same reason.

What operators should demand before signing

Founders and manufacturers considering industrial zone investment should treat the brochure as a starting document, not an answer.

First, test operational reliability. Obtain evidence on power, water, customs processing and the actual route from plant to port. “Facilitated” is not a service level.

Second, map people. Is there enough local talent for the first workforce and a plan to develop it? Can management and skilled staff live nearby with their families? A factory cannot scale on a labour plan that exists only in a presentation.

Third, examine governance. Are the zone’s foundational policies stable across administrations? Who enforces the rules? How are disputes handled? Investors can price high tax. They struggle to price arbitrary change.

Fourth, identify local suppliers before committing. A zone that imports every input and exports every finished product may create activity, but it creates a thinner local base than one that develops domestic procurement and capabilities.

Finally, separate the incentive from the operating case. Build the business model as if the tax saving were delayed or reduced. If the factory fails that test, the project is relying on a political feature to cover an economic weakness.

What serious capital will see

Experienced investors do not dislike incentives. They dislike being asked to treat them as a substitute for cash flows.

A lender or equity investor will want to know what happens when electricity fails, shipments are delayed or policy changes. They will ask whether the factory has customers, workable logistics and enough margin to survive normal disruption. They will also ask whether investment in the zone is additional, because a subsidised relocation is less durable than a genuinely competitive export business.

This is where first-time sponsors often miss the point. They present a tax holiday as revenue protection. The investor sees a benefit that depends on profitability, compliance and policy continuity. The investor then looks for the thing that creates profits in the first place: dependable operations.

It is the same discipline behind the question investors need answered before they commit. A persuasive pitch does not remove the objections that will appear in an investment committee paper. It merely delays them.

GI Network’s view: A bankable zone proposition begins with the cost and reliability of producing, clearing and moving goods. The incentive schedule belongs near the end of that analysis, not at the front.

In this situation, GI Network would test the proposed zone’s power, logistics, workforce, governance and supplier assumptions before investor outreach; model whether cash flows remain credible without heroic incentive assumptions; and prepare evidence that answers the objections capital providers will raise about additionality, policy stability and operating risk.

The Stay Test

Before backing a zone, use the Stay Test. It has five questions:

1. 1.Can the factory run? Test utilities and site services under normal operating conditions.
2. 2.Can the goods move? Test customs, port access and the road or rail corridor in real time, not on a map.
3. 3.Can the business staff itself? Test skills, training capacity and whether workers and families can live nearby.
4. 4.Can the rules be trusted? Test governance, enforcement and resilience to political change.
5. 5.Would the firm stay without the tax headline? Test supplier links, market access and the operational reasons that make relocation unattractive.

If a zone cannot answer the first four, the fifth is mostly theatre. If it can answer them, an incentive may be useful. But by then, the tax holiday is no longer the factory.