Mintz: Wrong Again on Corporate Taxes

Ten days ago, Jack Mintz released yet another paper claiming that international competitiveness requires continued corporate tax cuts. In addition to the usual questionable interpretations, it featured at least one straight factual error.

Mintz inaccurately reports Iceland’s 2010 statutory corporate tax rate as 15% (Table 2 on page 7 and Table 3 on page 9 in the PDF). In reality, Iceland raised its corporate tax rate back to 18% in 2010, as reported by the OECD, the World Bank, KPMG and PricewaterhouseCoopers.

I appreciate Mintz’s reluctance to acknowledge corporate tax increases, given his narrative that corporate taxes are always and everywhere falling. In fact, many countries plan to raise business taxes and toughen enforcement.

In Mintz’s simple average, Iceland counts for as much as the US. Still, a 3% error for one country out of 33 does not make much difference. The paper reports an average statutory rate for the OECD (excluding Estonia) of 25.7%. Iceland’s correct rate brings that average up to 25.8%.

However, the mistake is disconcerting since Mintz expects readers to take on faith his Marginal Effective Tax Rate (METR) calculations, which purport to integrate business tax provisions other than statutory rates.

No one can reproduce or scrutinize these calculations or the assumptions underlying them. Mintz runs the black box, the media reports the numbers it spits out, and policymakers listen intently.

Even if one accepts the METRs, a simple average of them is meaningless. If investment opportunities are roughly proportional to economic size, countries should be weighted by Gross Domestic Product.

For a few years – 2006, 2007 and 2008 – Mintz actually presented weighted averages. His problem was that they did not really make the case for further Canadian corporate tax cuts.

His solution was to stop presenting weighted averages. In 2009, 2010 and 2011, Mintz continued advocating corporate tax cuts based on simple averages alone.

As I demonstrated three weeks ago, Canada’s statutory corporate tax rate in 2010 was comfortably below the weighted OECD average. Applying the same approach to Mintz’s METRs reveals a weighted OECD average of 27.4%.

By comparison, Canada’s METR was just 20.5%. Federal and provincial governments could roll back corporate tax rates to 2010 levels or higher, while retaining a significant competitive advantage over other advanced economies as well as major emerging economies like Russia (31.9%), India (33.6%), Brazil (35.1%) and Argentina (43.1%).

UPDATE (March 9): Mintz has corrected Iceland.

METRs and GDP in the OECD
(GDP = expenditure approach, US$ billions, purchasing power parity)

 

2009

GDP

2010

METR

United States

14,044

34.6 %

Japan

4,135

29.5 %

Germany

2,975

23.8 %

France

2,173

34.0 %

United Kingdom

2,173

27.9 %

Italy

1,953

26.9 %

Mexico

1,540

17.5 %

Spain

1,481

25.4 %

Korea

1,321

29.5 %

Canada

1,276

20.5 %

Turkey

1,024

5.6 %

Australia

877

26.0 %

Poland

722

14.3 %

Netherlands

675

16.8 %

Belgium

392

(1.7 %)

Switzerland

350

17.6 %

Sweden

346

18.9 %

Greece

328

13.0 %

Austria

325

25.3 %

Norway

269

24.7 %

Czech Republic

268

12.0 %

Portugal

266

20.8 %

Chile

243

6.7 %

Denmark

208

18.5 %

Israel

206

14.6 %

Hungary

203

15.9 %

Finland

188

18.3 %

Ireland

177

10.9 %

New Zealand

125

17.6 %

Slovak Republic

124

11.2 %

Slovenia

56

11.6 %

Luxembourg

42

16.8 %

Estonia

27

N.A.

Iceland

12

8.9 %

9 comments

  • Does anyone know why the Belgian METR is negative? It seems hard to believe that Google (say) wouldn’t be running their money through Belgium instead of Ireland given that rate.

  • I think it’s mainly because Belgium allows corporations to deduct the assumed cost of equity financing (in addition to actual interest payments on debt financing.)

    This rate is marginal rather than average. It is not as though corporations get a tax refund on every dollar flowing through Belgium.

  • Thanks, Erin. So maybe I’m a little slow, but doesn’t that mean that, marginally, any company that has a sub in Belgium and a sub anywhere else, should try to run more and more business through the Belgian sub, as opposed to the other one? Maybe there’s a risk premium for the Flemish vs. Walloons unpleasantness.

  • My understanding of Mintz’s analysis is that, if an extra dollar of new investment in any OECD country generated the same pre-tax return for a corporation, it would invest the extra dollar in Belgium.

    I do not think that METRs have any bearing on the shifting of reported profits, which is why Mintz still examines statutory corporate income tax rates.

  • 1) Taxes are not the only factor affecting investment intentions. Workforce, net labour costs, infrastructure all play a part. However, it is notable that the European economy with the strongest growth (Germany) has a substantially lower METR than its nearest neighbours with the singular exception of Ireland.
    2) Richard Gordon in his blog has a strong discussion on tax incidence suggesting that corporate taxes do not land where most people think. http://worthwhile.typepad.com/worthwhile_canadian_initi/2010/08/incidence.html
    3) I would commend the Gentry study from the U.S. Treasury for people interested in a strong perspective on the issues of tax incidence.
    http://www.treasury.gov/resource-center/tax-policy/tax-analysis/Documents/ota101.pdf

  • Gordon’s views (and it is Stephen not Richard) are based on some flimsy evidence, rooted in a lot of bad theory.

    Here is what the Gentry study says. Tell me if you think this describes the world we live in:
    “While further research is necessary to draw definitive conclusions, these studies suggest that labor may bear a substantial burden from the corporate income tax. These empirical results are consistent with computable general equilibrium models based on an open economy in which a single country sets its tax policy independently of other countries; in these models, assumptions that capital is mobile and consumers are willing to substitute tradable goods produced in different countries imply that labor can bear more of the incidence of the corporate tax than capital bears.”

    CGE models are only quasi-empirical in that the results are driven by the assumptions made. I have yet to see an empirical result that definitely shows that changes in CIT rates translate into wage changes. This would be a really hard thing to prove, given all of the other things going on in the economy and labour market.

  • @ Marc,

    Yes a quasi-empirical empirical confirmation of a CGE pseudo reality. You can see why it might ring so true to an econometrician.

  • Sorry to pile on, but here is my response to a previous Stephen Gordon post on the same subject.

    By my count, Germany borders nine countries. Seven of them have substantially lower METRs, as calculated by Mintz: The Netherlands, Belgium, Luxembourg, Switzerland, the Czech Republic, Poland and Denmark. If anything, Germany’s relative economic success supports the view that business taxes are not decisive to investment and growth.

  • @ Erin

    Yah, but Ireland…..oh shit, never-mind.

Leave a Reply

Your email address will not be published.