Taxation and Economic Growth (OECD Working Paper 620)

Summary

"Taxation and Economic Growth" is OECD Economics Department Working Paper No. 620 (2008), by Åsa Johansson, Christopher Heady, Jens Arnold, Bert Brys, and Laura Vartia, all economists in the OECD's Economics Department at the time. It is the source of what has become the single most-cited piece of institutional evidence in tax-policy debates for shifting taxation away from income and toward property: the "tax and growth" ranking. A condensed, peer-reviewed version of the same research programme was later published as Jens Arnold, Bert Brys, Christopher Heady, Åsa Johansson, Cyrille Schwellnus & Laura Vartia, "Tax Policy for Economic Recovery and Growth," The Economic Journal, 121(550), F59–F80 (2011) — see research/arnold-tax-growth-ej. The working paper carries weight not because it is theoretically novel — the underlying claim that taxes on a fixed factor are less distortionary is textbook public finance — but because it is an OECD Secretariat cross-country empirical study, produced for and feeding directly into OECD tax-policy advice to member governments, and it has subsequently been cited across the IMF, World Bank, European Commission, and national tax-reform reviews (including the Mirrlees Review) as authoritative empirical backing for "growth-friendly tax structures."

The Core Argument and Findings

The working paper combines a survey of theory and prior evidence with new OECD empirical work at the macro, industry, and firm levels. The cross-country evidence behind the famous ranking comes, in the paper's own words (Box 11), from "a panel regression of GDP per capita covering 21 OECD countries over the period 1970 to 2005," with the econometrics set out in a companion working paper — Jens Arnold (2008), "Do Tax Structures Affect Aggregate Economic Growth? Empirical Evidence from a Panel of OECD Countries," OECD Economics Department Working Paper No. 643 — which uses annual data for 21 OECD countries over 1971–2004 and a Pooled Mean Group (PMG) error-correction estimator: short-run dynamics and country-specific convergence paths are allowed to differ across countries, while the long-run coefficients on the tax-structure variables are pooled. The overall tax-to-GDP ratio is included as a control, so the tax-share coefficients read as revenue-neutral shifts in the tax mix; on the level of taxation itself, the companion paper reports that "the estimations find a consistently negative coefficient for the overall tax burden" but cautions that "it would be premature to draw any policy conclusions from the sign of this coefficient." On the mix, WP 620's summary finding is that "[t]axing consumption and property appears to have significantly less adverse effects on GDP than taxing income."

Disaggregating tax revenue by type, the paper's central and most widely cited result is a ranking of tax categories by estimated harm to long-run growth, from most to least harmful:

1. Corporate income taxes — most harmful to growth
2. Personal income taxes
3. Consumption taxes (e.g. VAT)
4. Recurrent taxes on immovable property — least harmful to growth

In the abstract's own words: "Corporate taxes are found to be most harmful for growth, followed by personal income taxes, and then consumption taxes. Recurrent taxes on immovable property appear to have the least impact." The paper's summary of policy options states the same ranking positively: the evidence "suggests a 'tax and growth ranking' with recurrent taxes on immovable property being the least distortive tax instrument in terms of reducing long-run GDP per capita, followed by consumption taxes (and other property taxes), personal income taxes and corporate income taxes" (para. 9). The abstract concludes that "[a] revenue neutral growth-oriented tax reform would, therefore, be to shift part of the revenue base from income taxes to less distortive taxes such as recurrent taxes on immovable property or consumption." On magnitudes, Box 11 reports that "a shift of 1% of tax revenues from income taxes to consumption and property taxes would increase GDP per capita by between a quarter of a percentage point and one percentage point in the long run depending on the empirical specification," while warning in the same breath that "[t]he magnitude of the estimated effect is larger than what would be reasonably expected" and "should be interpreted with caution." This ranking is the empirical basis for the widely repeated claim that property taxes are the "most growth-friendly" or "least distortive" tax in the OECD's toolkit — a claim that has since propagated through OECD Going for Growth reports, IMF and World Bank tax-policy guidance, and numerous national tax reviews.

Relation to the Georgist Case

This paper is frequently invoked as institutional confirmation of the Georgist claim that taxing land/property is uniquely non-distortionary, and it supports the broader outcome that land value tax can substitute for capital taxation without an efficiency loss: the paper's own ranking puts recurrent property taxation at the opposite end of the harm spectrum from corporate income tax, the tax on capital it is most often proposed to replace. Because the estimates come from realised, decades-long cross-country data on actual OECD tax systems rather than a theoretical model, the paper is frequently treated as harder, "real-world" evidence for the same conclusion that the deadweight loss theory predicts on first principles: a tax on an immobile, inelastically supplied base does less economic damage than a tax on a mobile or elastically supplied one.

That said, the paper's policy relevance to land value taxation specifically is a step removed from what it actually measures — see Nuances and Limits below.

Nuances and Limits

Several caveats are essential to using this paper honestly in the Georgist case:

• "Recurrent taxes on immovable property" is not land value tax. The OECD Revenue Statistics category the paper analyses (OECD tax category 4100, "recurrent taxes on immovable property") includes ordinary property taxes levied on the combined value of land and buildings/improvements — the actual property tax regimes of the countries in the panel, almost none of which used a pure land value tax base at the time. The efficiency argument specific to land (as opposed to structures) is not separately tested; the paper cannot distinguish how much of property tax's favourable ranking derives from the land component versus the improvements component. This is the same category-conflation flagged on this wiki's LVT is just a property tax objection page — the OECD's finding is evidence for property taxation broadly, and only suggestively, not directly, for a land-only base.
• Correlation and identification. A PMG panel regression can support a well-identified long-run association, but causal identification in cross-country macro-growth regressions is inherently difficult: reverse causality (richer, faster-growing countries may choose different tax mixes for other reasons) and omitted country-specific factors are standard concerns for this literature, and the authors' claims should be read as "associated with" rather than proven causal effects.
• Robustness of the ranking has been directly challenged. Jing Xing, "Tax structure and growth: How robust is the empirical evidence?", Economics Letters 117(1), 2012, 379–382, re-examines the PMG specification underlying the ranking. Per the published abstract, "the 'tax and growth ranking' suggested by some recent empirical studies is not robust under different assumptions about heterogeneity across countries of the long-run and short-run coefficients in the underlying econometric model," because "[e]vidence for significant tax structure effects depends on long-run parameter homogeneity restrictions, underlying pooled mean group estimation, which are found to be invalid." On the specific question of whether this disturbs the property-tax result: in the fuller working-paper version of the same study (Jing Xing, "Does tax structure affect economic growth? Empirical evidence from OECD countries," Oxford University Centre for Business Taxation WP 11/20, 2011, using 17 OECD countries over 1970–2004), the property result is the one finding that survives — "we do not obtain compelling evidence favouring consumption taxes over income taxes, or favouring personal income taxes over corporate income taxes. The only robust result appears to be that shifts in tax revenue towards property taxes are associated with a higher level of income per capita in the long run." Xing's re-estimation thus mainly overturns the ordering among consumption, personal income, and corporate taxes; the pro-property-tax finding is the most stable at the aggregate level. One disaggregated caveat (WP 11/20, Column 11, also carried on the EJ article's page): when property taxes are split into sub-categories, it is not the recurrent-immovable-property sub-category that robustly tops the ranking — the "other property taxes" coefficient is significantly larger — and the published note's abstract frames tax-structure significance in general as resting on the invalid pooling restrictions.
• Later replication is more skeptical of the whole tax-shift claim. Donatella Baiardi, Paola Profeta, Riccardo Puglisi & Simona Scabrosetti, "Tax policy and economic growth: does it really matter?", International Tax and Public Finance 26(2), 2019 (working paper version: CESifo WP 6343, 2017), revisit the direct-to-indirect tax shift question with an extended sample and time period and find no robust relationship between revenue-neutral tax shifts and growth, casting doubt on the growth-enhancing effects of shifting from direct to indirect taxation that the OECD's 2008 result implied. Their paper does report a negative association between total tax revenue and growth, consistent with part of the original finding, but is considerably more cautious about the tax-composition result that underlies the "tax and growth ranking."
• Scope is limited to OECD, high-income countries over 1971–2004. The ranking's applicability to developing countries with different administrative capacity, informal sectors, and tax structures is not established by this paper; see World Bank property tax determinants and property tax raises welfare in developing countries for evidence more specific to that context.
• The paper does not evaluate a specific LVT reform proposal. It is a positive, descriptive empirical study of historical tax structures and growth, not a policy simulation of what would happen if a country replaced its corporate tax with a land value tax; the "replace capital taxes with land taxes" policy conclusion is an inference commonly drawn from the ranking, not a claim the paper itself tests directly.

Bears On

• Outcome: LVT can replace capital taxes without efficiency loss — the paper's ranking places corporate tax as most harmful and recurrent property tax as least harmful, the most-cited cross-country empirical support for shifting tax burden from capital toward property/land.
• Objection: LVT is just a property tax — this paper is itself a case study in the conflation the objection describes: its "least harmful" category is ordinary property tax (land + improvements), not a land-only base.
• Concept: Deadweight Loss — the paper's empirical ranking is commonly read as real-world confirmation of the theoretical deadweight-loss argument for taxing inelastically supplied bases.
• Research: The Mirrlees Review — cites this OECD tax-and-growth literature as part of the mainstream evidentiary basis for recommending a shift toward land/property taxation.
• Research: research/arnold-tax-growth-ej — the peer-reviewed Economic Journal article drawing on the same underlying research programme and estimates.

See Also

• The Mirrlees Review: Tax by Design
• research/arnold-tax-growth-ej
• Deadweight Loss
• Objection: LVT is just a property tax
• Taxing Immovable Property: Revenue Potential and Implementation Challenges
• World Bank property tax determinants

Sources

1. Åsa Johansson, Christopher Heady, Jens Arnold, Bert Brys & Laura Vartia (2008), "Taxation and Economic Growth," OECD Economics Department Working Papers, No. 620, OECD Publishing, DOI: 10.1787/241216205486; issued as OECD document ECO/WKP(2008)28, 11 July 2008 (cover title "Tax and Economic Growth"). OECD · free full-text PDF — the full primary PDF was fetched and read directly (via the Internet Archive's copy of the one.oecd.org PDF); used for the abstract, the tax-and-growth ranking and policy-conclusion quotations, and the Box 11 panel details and magnitude caveats. All direct quotations on this page are verbatim from the primary PDF.
2. Jens Arnold (2008), "Do Tax Structures Affect Aggregate Economic Growth? Empirical Evidence from a Panel of OECD Countries," OECD Economics Department Working Papers, No. 643, OECD Publishing, DOI: 10.1787/236001777843; issued as ECO/WKP(2008)51. Free full-text PDF — the companion econometric paper that WP 620's Box 11 defers to "for details"; used (read directly) for the PMG error-correction estimator, the 21-country 1971–2004 sample, the ranking as stated in its abstract, and the caution against interpreting the negative overall-tax-burden coefficient.
3. Jens Arnold, Bert Brys, Christopher Heady, Åsa Johansson, Cyrille Schwellnus & Laura Vartia (2011), "Tax Policy for Economic Recovery and Growth," The Economic Journal, 121(550), F59–F80, DOI: 10.1111/j.1468-0297.2010.02415.x. Wiley — used to establish the peer-reviewed companion publication of the same research programme (see research/arnold-tax-growth-ej).
4. Jing Xing (2012), "Tax structure and growth: How robust is the empirical evidence?", Economics Letters, 117(1), 379–382, DOI: 10.1016/j.econlet.2012.05.054. ScienceDirect · abstract via IDEAS/RePEc — used for the robustness critique; the published abstract is quoted verbatim (cross-checked against the identical abstract on the Oxford CBT page).
5. Jing Xing (2011), "Does tax structure affect economic growth? Empirical evidence from OECD countries," Oxford University Centre for Business Taxation Working Paper 11/20. Free full-text PDF — the fuller working-paper version of Xing (2012), read directly; used for the finding that the property-tax shift is the only robust result.
6. Donatella Baiardi, Paola Profeta, Riccardo Puglisi & Simona Scabrosetti (2019), "Tax policy and economic growth: does it really matter?", International Tax and Public Finance, 26(2), 282–316. SSRN — used for the later, more skeptical replication of the direct-to-indirect tax shift and growth claim.
7. Institute for Fiscal Studies, Tax by Design: The Mirrlees Review (2011). IFS — used to establish this paper's downstream institutional influence; see research/mirrlees-review.