Risks and Tradeoffs

A serious strategy has to engage seriously with its strongest critics. The interventions cataloged here have four main lines of objection: that they will cause capital flight and reduce investment; that they will reduce innovation and entrepreneurship; that they require state capacity that many countries lack; and that aggressive redistribution can produce democratic backlash of the kind that empowered populist movements in the 2010s. Each objection has empirical content worth taking seriously, and the strategy’s design responds to each.

The capital flight objection is the most testable. The empirical evidence (Kleven et al. 2020; Akcigit et al. 2022; the Scandinavian wealth-tax retreats) shows that high-tail elasticities are real but smaller than feared and are largely a function of coordination failure — which is why Global Tax Coordination is a pillar of the strategy rather than an afterthought. The innovation objection draws on cross-country growth studies (Jones & Kim 2018; Aghion et al. 2019) that find mixed effects of top tax rates on innovation, with the dominant effect coming through the type of activity taxed (rent-seeking vs. productive) rather than the level.

The state capacity objection is the most serious for global application. Many of these interventions presuppose functioning tax administrations, independent regulators, and a free press — which is exactly what the wealth-concentration feedback loop has eroded in many countries. Building or rebuilding state capacity is therefore an implicit precondition. The backlash risk is real and contested: some evidence (Ferrari et al. 2023; Mounk 2018) suggests redistributive failure breeds populism, while other evidence (Berman 2019; Eichengreen 2018) suggests redistributive success breeds durable centrist coalitions.

Objection matrix

Objection	Strongest evidence	Strategy’s response
Capital flight	Scandinavian wealth-tax exits, Kleven et al.	Global coordination as a pillar; design for low elasticity (inheritance, immobile assets)
Innovation harm	Aghion et al. mixed; Akcigit et al. mixed	Tax rents heavier than productive returns; preserve R&D credits
State capacity	Tax-administration weakness in LMICs	Sequence: transparency first; partner with OECD and UN technical assistance
Democratic backlash	Right-populist surges in U.S., U.K., Italy, Brazil	Predistribution + universalism rather than means-tested redistribution
Brain drain	Modest in the data (Akcigit et al.)	International coordination + targeted residency rules
Compliance cost	Real for SMEs	Thresholds, simplified regimes for small entities
Black-market growth	Real for high marginal rates	Third-party reporting, automatic information exchange

The capital-flight literature in depth

The capital-flight objection has been the central empirical battleground for the strategy in this wiki. The strongest version of the objection is that any meaningful increase in top-end taxation will produce migration of the wealthy and their capital to lower-tax jurisdictions, hollowing out the tax base and producing little net revenue.

The empirical record now allows reasonably confident quantitative answers. Three key papers:

Kleven, Landais & Saez (2013, AER, “Taxation and International Migration of Superstars”) studied European football players’ migration responses to country-level top tax rates and found substantial elasticities (~1) for foreign players, but much smaller (~0.15) for domestic players. The combined elasticity for any given country’s labor force is in the range 0.2–0.4 — meaningful but not catastrophic.

Akcigit, Baslandze & Stantcheva (2016, AER, “Taxation and the International Mobility of Inventors”) studied superstar inventors’ migration responses and found similar elasticities for foreign inventors (~1) and small elasticities (~0.03) for domestic inventors. Crucially, the migration response is concentrated in foreigners considering relocation, not in citizens leaving their country of origin.

Jakobsen, Kleven, Kolsrud, Landais & Munoz (2024, NBER WP 32153, “Taxing Top Wealth: Migration Responses and their Aggregate Economic Implications”) is the most recent and most directly relevant. Using Scandinavian wealth-tax variation, they find that a 1 percentage point increase in the top wealth tax rate reduces the stock of wealthy taxpayers by 1.76%, with associated reductions of 0.05% in aggregate employment, 0.07% in aggregate investment, and 0.13% in aggregate value-added. These are measurable but small — the predicted catastrophe does not materialize at the scale of historically observed wealth-tax changes.

Advani, Burgherr & Summers (2025, U.K. evidence) is U.K.-specific and finds similarly modest migration responses to recent U.K. top-rate changes.

The strategy’s response to the capital-flight literature has three components:

1. Most of the elasticity is foreign, not domestic. Tax design that exempts new immigrants for an initial period (as several European countries do) substantially reduces the binding constraint without generating substantial revenue loss.
2. Coordination dramatically reduces the elasticity, because the threat of relocation is only credible when there are competitive lower-tax destinations. This is the Global Tax Coordination argument.
3. Some assets are intrinsically immobile — real estate, in-country business operations, family ties — and the most progressive instruments (estate taxes, capital gains at realization, mark-to-market for liquid securities) operate disproportionately on these immobile bases.

The honest summary is that capital flight is real but smaller than the policy debate assumes, and design choices substantially determine its magnitude.

The “type of activity” framing of the innovation question

A more sophisticated version of the capital-flight argument is that high top tax rates damage innovation — productive entrepreneurship, R&D investment, the long-run dynamism of the economy. This is a more serious objection because it concerns growth rather than revenue, and growth concerns trump revenue concerns in most political settings.

Aghion, Akcigit, Bergeaud, Blundell & Hémous (2019, RES, “Innovation and Top Income Inequality”) is the most-cited recent study. The authors find that innovation (measured by patents per capita) accounts for ~17% of the U.S. top 1% income share rise between 1975 and 2010. They use cross-state and IV variation to argue for a causal direction running from innovation to top-end inequality, with implications for the design of progressive taxation.

The headline finding has two important refinements:

• Innovation does not increase broader inequality (Gini, Palma) — the effect is concentrated specifically at the top tail.
• Innovation is positively correlated with social mobility in most settings — but the correlation weakens substantially in states with high lobbying activity. This is consistent with the rent-seeking-vs-innovation distinction.

Akcigit, Grigsby, Nicholas & Stantcheva (2021, QJE, “Taxation and Innovation in the Twentieth Century”) found that higher personal income taxes negatively affect the quantity of innovation (patent counts) but with mixed effects on innovation quality. The magnitude of the disincentive is meaningful but not catastrophic — comparable in size to the elasticity of taxable income.

The strategy’s response is the type-of-activity framing:

• Tax rents heavier than productive returns. Inheritance taxes, monopoly-rent capture (antitrust enforcement), and high marginal taxes on capital gains from non-productive sources hit unproductive activity disproportionately.
• Preserve R&D credits and direct innovation subsidies. The U.S. R&D tax credit and equivalent OECD instruments largely insulate productive innovation from the tax-rate disincentive.
• Distinguish founding from inheriting. The largest concentration of unproductive top-end wealth is inherited, not founder-generated. Estate-tax reform addresses this directly without affecting the founder-generation incentive.

The honest summary is that innovation effects are real but design-sensitive, and a well-designed progressive tax architecture can target unproductive activity disproportionately while preserving (or enhancing) productive innovation.

State capacity as the binding constraint

The most serious objection to the global application of the strategy is state capacity. Each pillar in the architecture presupposes a functioning tax administration, independent regulators, a capable judiciary, free media, and credible enforcement. These are exactly what the wealth-power feedback loop has eroded in many countries.

The constraint is most binding in:

• Low-income countries with limited tax-administration capacity (most of sub-Saharan Africa, parts of South Asia).
• High-corruption middle-income countries where regulatory and enforcement institutions exist on paper but are systematically captured (Russia, Nigeria, several Latin American cases).
• Countries with eroded democratic institutions where independent regulators have been hollowed out (Hungary, Turkey, increasingly several others).

The strategy’s response is sequencing-based. State capacity has to be built before the substantive interventions can operate. The implication is that:

1. Beneficial-ownership transparency is the highest-priority intervention in low-state-capacity contexts because it can be implemented with international technical assistance and produces measurable revenue gains.
2. OECD/UN technical assistance programs (IMF Fiscal Affairs Department, OECD CTPA, ATAF, CIAT) are the operational vehicles for state-capacity building in tax administration.
3. International coordination can substitute partially for domestic enforcement weakness — Pillar Two top-up tax can be collected by a stronger jurisdiction even if the source country cannot enforce its own minimum.

The state-capacity objection cannot be talked away. Honest application of the strategy in low-capacity contexts requires substantial international investment in capacity-building infrastructure that does not currently exist at scale. This is one of the strongest arguments for the UN Framework Convention on International Tax Cooperation, which would create a multilateral architecture for capacity-building alongside its rule-writing function.

The populism literature

The fourth substantive objection is that aggressive redistribution can produce democratic backlash — populist movements that exploit the disruption of redistribution to advance authoritarian or nativist agendas. The 2010s populist surge in the U.S., U.K., Italy, Brazil, and elsewhere has often been read as evidence for this risk.

The empirical literature is genuinely contested. Two major positions:

The redistribution-failure-breeds-populism view. Yascha Mounk’s The People vs. Democracy (Harvard 2018) and Larry Bartels’ Democracy Erodes from the Top (Princeton 2023) argue that the 2010s populist surge was largely a response to the failure of mainstream politics to deliver on redistribution after the 2008 crisis. Voters who experienced sustained material decline turned to outsider candidates regardless of those candidates’ actual policy positions. The implication is that more aggressive redistribution would have prevented populism, not produced it.

The redistribution-itself-breeds-backlash view. A different reading (Sheri Berman, Democracy and Dictatorship in Europe, 2019; some interpretations of Eichengreen) emphasizes the cultural sources of populism — immigration anxiety, status loss, identity politics — and argues that economic redistribution is largely orthogonal to these drivers. On this reading, redistribution policies do not cause backlash, but they do not prevent it either.

The strategy’s response is to design redistributive policies that are maximally universalist and minimally means-tested, on the theory that universal benefits build broader political coalitions and avoid the “deserving vs. undeserving” framing that populism exploits. Predistribution interventions (UBI, baby bonds, sovereign-wealth dividends) are particularly favored on this margin. The empirical question of whether universal vs. means-tested design substantially affects populism risk is not settled.

A separate concern is the specific reaction to visible wealth taxation. Wealth-tax debates in France, Spain, and Norway have produced substantial elite political mobilization that has at times allied with broader anti-government movements. This is a real tactical risk that wealth-tax design has to anticipate.

What the strategy gives up

A strategy is partly defined by what it does not prioritize. The four-pillar architecture in this wiki gives up several alternative approaches that have legitimate proponents:

It gives up speed. A more aggressive strategy (immediate large wealth taxes; immediate breakups of dominant firms) might produce larger short-term inequality reductions than the incremental sequencing here. The cost of incrementalism is that the next 10–15 years of inequality dynamics will continue largely unchanged while the foundation pillars are built.

It gives up the upper-bound effect size. Tier 3 interventions have larger expected effects than Tier 1 interventions; prioritizing Tier 1 because of higher tractability means accepting smaller effects in exchange for higher probability. A more risk-tolerant strategy would weight Tier 3 more heavily.

It gives up some forms of universalism. Universal Basic Income at meaningful scale would be a more thoroughgoing universalist intervention than the predistribution mix recommended here. The strategy’s preference for asset-building (baby bonds, employee ownership) over income flooring (UBI) reflects a judgment that asset-building has more durable political and economic effects, but this judgment is contested.

It gives up movement-political emphasis. The strategy here is technocratic and institutional. A more movement-oriented strategy would emphasize organizing, mobilization, and electoral realignment over the policy designs and institutional infrastructure described here. Both approaches are necessary; the relative weight is genuinely debated.

It gives up some country-specificity. The strategy is broadly applicable to the high-income democracies and the middle-income countries with substantial concentrated wealth. It is less directly applicable to low-income countries with very different starting conditions and to countries with already-substantial existing pillars (Norway, Sweden in some respects).

The broader epistemic point is that no single strategy can be optimal across all conditions and goals. The four-pillar architecture here is a defensible strategy, not a uniquely correct one. Reasonable people working in this space will weight the trade-offs differently and design different approaches. The wiki’s purpose is to make the underlying empirical and structural arguments visible enough that those alternative approaches can be evaluated on their own merits.