Start with what a government actually offers. In Kenya, a developer who builds a hotel can deduct half the cost of the building from taxable income in the first year of use, and a quarter of the remaining value each year after that, under the investment allowance of the Income Tax Act. Tourism packages are zero-rated for value-added tax. Tour vehicles, sight-seeing buses and tourism boats can be imported free of customs duty. A hotel established inside a Special Economic Zone sits outside the customs territory altogether, exempt from import duty, VAT and excise, and pays a reduced rate of withholding tax on payments to non-residents. Losses can be carried forward for ten years against future profit. None of this is hidden or improper. It is the published, standard incentive package that the Kenyan state, through its own investment-promotion agency and revenue authority, offers to attract investment into hotels.1 Versions of it exist across the continent. The question this issue asks is a simple one that the evidence answers with unusual consistency: does the incentive actually cause the hotel to be built, or does it hand public money to a building that was coming anyway.
The Corridor Index framework reads a tourism economy by tracing where the value is captured and who controls the terms. Across this series the value has flowed to a foreign platform, a foreign brand and a foreign carbon buyer, each capturing a share of what African tourism generates. Tax incentives belong to the same story, but they invert its direction. Here the state is not losing value to an outside actor that took it. The state is giving it up, in advance, by choice, in the hope of attracting investment. That makes this the sharpest test of economic sovereignty in the whole series, because taxation is the part of sovereignty a government controls most completely. No platform, no brand and no currency peg stands between a state and its own tax code. When value leaks here, it leaks through a decision the state made itself.
What the evidence says about whether it works
The evidence on tax incentives is one of the more settled findings in development economics, and it is not encouraging for the states that rely on them. The International Monetary Fund, reviewing the question for low-income countries, has found that tax incentives generally rank low in the factors that actually drive investment decisions, well below political stability, infrastructure, the cost and skill of labour, and the size of the market. A study of the CFA Franc zone in Sub-Saharan Africa found no reliable positive relationship between tax holidays and investment. A broader analysis found that extending a tax holiday by ten years raised foreign direct investment by only about one percentage point of GDP, and that incentives are up to eight times less effective in weak investment climates than in strong ones, which places most of Africa in the category where they work least.2 The pattern that emerges is consistent across studies. Investors decide where to build on the fundamentals, and take the tax holiday as a bonus where it is offered. The incentive rarely tips the decision. It mostly discounts an investment that was already going to happen.
The clearest illustration comes from the firms themselves. Research by UNU-WIDER on corporate tax holidays found that the majority of firms receiving them would have remained profitable without them, which means the holiday did not rescue a marginal project but subsidised a viable one. The same study found the incentive granting the top firms a net profit margin near ninety percent where without it they would have sat around sixty. That is the mechanism in miniature. The tax holiday did not conjure the investment. It widened the margin of an investment that was already sound, at the public's expense.3
The cost, which is real and quantified
Because the incentive is a tax not collected rather than a cheque written, its cost is easy to lose. Governments that would scrutinise a line of spending will wave through an exemption of equal size, because the exemption never appears as spending. The technical term is tax expenditure, and where states have measured it the numbers are large. South Africa's National Treasury has put the revenue forgone through incentives at around 4.5 percent of a tax base of roughly 1.3 trillion rand, which is tens of billions of rand each year that the state chose not to collect. In Uganda, the conservative estimate of revenue lost to a handful of studied tax holidays reached close to a fifth of the country's total corporate income tax in the peak year. In Nigeria, two subsidiaries of a single group saved more than 280 million dollars over five years under one holiday scheme.4 These are not rounding errors in national accounts that are already stretched. They are hospitals not built, teachers not paid, and debt serviced with borrowing rather than revenue, given up in exchange for investment the evidence says would largely have arrived regardless.
The race that no one wins
The reason states keep offering incentives the evidence says do not work is that they are afraid of what happens if their neighbour offers them and they do not. This is the trap, and it is well documented. The IMF has found that the length of a country's tax holidays responds positively to the length of its neighbours', meaning states lengthen their giveaways in reaction to one another rather than to any assessment of what works. Across special regimes in Africa, effective tax rates in the favoured sectors have fallen to almost zero. Each government, acting rationally to avoid losing an investor to the country next door, joins a competition that leaves every participant with less revenue and roughly the same investment they would have had if none of them had started. Tourism is especially exposed to this, because a hotel investor genuinely can choose between coastlines, and every ministry knows it. The result is a continent of neighbours bidding down the price of their own tax base against each other, to the benefit of investors who play them off and would, on the evidence, have come to one of them regardless.5
Each government, acting rationally to avoid losing an investor to the country next door, joins a competition that leaves every participant with less revenue and roughly the same investment. It is the one leak in the series the continent inflicts on itself.
The honest limit of the argument
The case against incentives can be pushed further than the evidence supports, and two cautions keep it honest. The first is that not all incentives are equal, and the research that condemns tax holidays is kinder to others. Cost-based incentives, which reward money actually invested in buildings, plant and equipment, perform better than profit-based holidays, because they are tied to real capital formation rather than to profit that may have existed anyway; Kenya's investment deduction is closer to this better category than to a naked holiday. The literature also finds that legal guarantees for investors and a simpler, more predictable tax system attract investment more reliably than any specific giveaway, which points to reform rather than pure removal. The second caution is that a state genuinely trying to build a tourism sector from a weak base may reasonably use a time-limited, well-targeted incentive to signal seriousness and lower the risk of a first mover, and there are cases where that has helped. The Corridor's claim is the specific one that survives both points: the broad, profit-based, competitively-driven incentives that dominate African tourism promotion forgo revenue the continent cannot spare, for investment the evidence says would mostly have come anyway, and a government that offers them without measuring their cost against their effect is giving away sovereignty it was under no obligation to surrender.
Three tests over the next year
The first test is measurement. Whether more African states publish a tax expenditure budget, putting a number on the revenue their tourism and investment incentives forgo, will determine whether these giveaways are ever weighed as the spending they are, since an exemption that is never counted is never questioned. The second is design. Whether states shift from profit-based holidays toward cost-based deductions tied to real investment, and toward the legal certainty and administrative simplicity the evidence favours, would let a government support tourism investment without simply discounting profit that needed no help. The third is coordination. The same regional bodies that this series has watched struggle to open borders and skies could, in principle, agree a floor beneath tax competition, a common minimum that stops African neighbours bidding their own revenue down against one another; the global minimum tax now taking shape gives them a template. Africa gives away this value by choice. Whether it decides to stop is a question of measurement, design and the willingness to coordinate, and unlike the other leaks in this series, it is entirely within the continent's own hands.