Tax Competition and International Taxation - PowerPoint PPT Presentation


PPT – Tax Competition and International Taxation PowerPoint presentation | free to download - id: 4afd1c-MzBjO


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation

Tax Competition and International Taxation


Tax Competition and International Taxation George R. Zodrow Professor of Economics Rice Scholar, Baker Institute for Public Policy Rice University – PowerPoint PPT presentation

Number of Views:266
Avg rating:3.0/5.0
Slides: 89
Provided by: MHen5


Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Tax Competition and International Taxation

Tax Competition and International Taxation
  • George R. Zodrow
  • Professor of Economics
  • Rice Scholar, Baker Institute for Public Policy
  • Rice University

  • Basic Tax Competition Models Primarily METRs
  • Extension to Statutory Rates, AETRs
  • Other Extensions
  • Five Major Qualifications
  • Empirical Evidence on Tax Competition
  • Conclusion

Basic Tax Competition Models
  • Simplest models assume taxing jurisdiction (state
    in a federation, nation in world economy) is
    small open economy fixed r and fixed p for
    tradable goods
  • Cournot-Nash competition in tax rates across
    jurisdictions competitive firms
  • Assuming tax on local factors available,
    source-based tax on internationally mobile
    capital is undesirable the zero tax result
    (Zodrow-Mieszkowski, 1983, 1986 Gordon, 1986
    Razin-Sadka, 1991)

  • Logic underlying the zero tax result
  • Source-based tax on capital income cannot be
    borne by perfectly mobile capital
  • Borne instead by local residents (lower wages,
    land rents higher prices for non-tradable goods)
  • Indeed, Harberger (2008) argues that burden of
    corporate income tax borne more than 100 by
  • Labor in tradable good corporate sector bears all
  • But labor mobile across sectors, wages decline
  • So, all labor bears gt100 (he estimates 130)

  • Accordingly, may as well tax local residents
    directly and at least avoid efficiency costs of
    taxing mobile capital due to
  • Emigration of capital
  • Lower K-intensity in production
  • Tax bias against capital-intensive goods
  • Inefficient product specialization (Wilson, 1987)

  • Another potential source of economic inefficiency
    is public service underprovision
  • Suppose government constrained to apply
    source-based tax on capital income, e.g., because
  • Must apply same property tax rates to residential
    and non-residential property, or must apply same
    corporate tax to both domestic and foreign firms
  • Wants to tax economic rents, especially
    location-specific rents

  • Need backstop for individual income tax
  • Political constraints favoring some form of
    business taxation, especially of foreigners
  • Then government tends to under-provide public
  • Government reduces tax to reduce tax-induced
    capital emigration (Zodrow-Mieszkowski, 1986
    Wilson, 1986)
  • Degree of inefficiency may be large (Wildasin,
  • Included in under-provision of government
    services reduction in social safety net (Sinn,
    1994, 1996, 1997)

  • Both models theoretically can generate
    race-to-the-bottom in capital income taxation
  • Small open economies move toward zero tax
  • As constraints on use of capital income taxes
    relaxed in tax competition model, get lower K
  • Note Under-provision result does not extend to
  • Benefit taxes on mobile capital
  • Taxes on external costs imposed by capital

  • Models with significant spillovers of public
  • With spillovers across jurisdictions, attracting
    K from other jurisdictions is less desirable,
    since lowers their G
  • With pure public good, no under-provision
    (Bjorvatn-Schjelderup, 2002)
  • Models with rents earned by foreigners
  • Tendency for under-provision limited by tendency
    for over-provision due to tax exporting potential
    (Wagner-Eijffinger, 2005 Mintz, 1994
    Huizinga-Neilsen, 1997)
  • Sorensen (2006) still under-provision big rate
    effect (drops 11 points if foreign share drops
    25 to zero)

Tax Competition Statutory Rates
  • Logic of basic tax competition models applies
    primarily to METRs (CRS production functions, PC)
  • Suggests that cash flow tax with high statutory
    rate but METR0 would not be problematical
  • But more recent models suggest that tax
    competition extends to statutory rates as well,
    which may even be more important, due to
  • 1. Competition for firms with firm-specific rents
  • 2. Potential for income shifting

  • 1. Firm-specific rents
  • Many MNCs earn firm-specific rents rents seem to
    be more important in recent years (Auerbach,
  • Indeed, such rents are prime rationale for MNCs
  • unique technological knowledge
  • superior managerial skills or production
  • valuable brands, trademarks, reputations
  • other intangible assets

  • But MNCs with firm-specific rents are also highly
    mobile internationally and especially prized by
    national governments due to external benefits
  • Key tax rate to attract such firms is not METR
    but statutory tax rate or AETR that applies to
    economic rents
  • Especially if firm making discrete location
    choice, e.g., to maximize economies of scale to
    serve multiple markets (Devereux, et al.)
  • So, tax competition also applies to statutory
    rates (Gordon and Hines, 2002)

  • 2. Income shifting
  • MNCs can reallocate profits among countries
  • Transfer prices
  • Loans, other transfers between subsidiaries
  • Allocate general expenses
  • Also puts downward pressure on statutory rates
    STR is prime determinant of income shifting
  • Low tax rate attracts revenues, repels deductions
  • Attracts firms as facilitates use of avoidance

  • Indeed, tax competition over STRs due to
    potential for income shifting or appropriating
    location-specific rents may be more important
    than standard tax competition over METRs to
    attract mobile capital
  • Haufler and Schjelderup (2000) with income
    shifting potential (transfer pricing), can be
    desirable to lower statutory rate, even if raises
    METR on new I, especially w/ some foreign owners

  • Fuest and Hemmelgarn (2005) similar result if
    firms can easily reallocate debt
  • Becker and Fuest (2005) with differentially
    mobile firms and differential profit rates,
    desirable to lower rates and raise METRs if
    mobile firm has relatively high profits
  • Devereux, Lockwood and Redoano (2008)
  • Construct model with tax competition for mobile
    capital (METR) and profit shifting (STR)
  • Also get similar result Taxing country may
    respond to competitors lower statutory rates
    with lower own statutory rate but higher METR

  • Much empirical evidence confirms income shifting
  • Profits tend to be disproportionately reported in
    low-tax countries (Hines, 1999)
  • All observed differences in profitability across
    countries are due to allocations of intangible
    income and debt (Grubert, 2003)
  • Revenues due to unilateral tax increase are
    offset by 65 due to additional use of transfer
    pricing only (Bartlesman and Beetsma, 2003)

  • Deductible interest payments are concentrated in
    high-tax countries while non-deductible dividends
    payments are concentrated in low-tax countries
    (Altshuler-Grubert, 2002 Hines-Hubbard, 1990
    Grubert, 1998)
  • Location of debt depends on tax rate differences
    (Desai-Foley-Hines, 2004 Huizinga-Laeven-Nicodeme
    , 2006)
  • Inter-firm loans used to reduce tax liability by
    German firms (Buettner and Wamser, 2007)

  • Deductible royalties substituted for
    non-deductible dividends in high-tax countries
    (Grubert, Randolph and Rousslang, 1996 Grubert,
  • Location of RD and other intangibles is highly
    tax sensitive (Hines, 1995 Altshuler-Grubert,
    2004 Mutti-Gruber, 2006)
  • Foreign affiliates with nearby tax haven pay
    lower taxes (Desai-Foley-Hines, 2006)

  • Bosworth, et al. (2007) 1/3 of excess returns
    earned by US FDI (relative to foreign FDI in US),
    due to income shifting to low-tax countries
  • And, income shifting is increasing over time
    (Altshuler-Grubert, 2006)

  • Similar results in Canada
  • Income shifting within provinces occurs in
    response to tax differentials (Mintz-Smart, 2004)
  • Debt reallocation across Canada and the US in
    response to tax differentials favoring US in mid-
    1980s (Jog and Tang, 2001) estimate large
    effects on Canadian revenues

Arguments May Extend to Subsidization
  • Imperfect competition
  • Subsidy may be desirable to offset too-high
    factor prices if monopoly power in capital
    markets and monopoly profits taxed (Judd, 1997,
  • Argues this is especially true in equipment
  • Imperfect information
  • Subsidy may be desirable to offset foreigners
    poor information on investment prospects, market
    conditions, accounting rules, etc. (Gordon and
    Bovenberg, 1996)

Other Extensions to Basic Model
  • Many extensions to the basic tax competition
    model (described in Wilson, 1999 Fuest, Huber
    and Mintz, 2003 Zodrow, 2003 Wildasin and
    Wilson, 2004)
  • General tax competition/under-provision results
    have been extended to
  • Labor income taxes (rather than head taxes) as
    alternative tax with variable labor supply
    (Bucovetsky and Wilson, 1991)
  • Labor mobility across jurisdictions (Brueckner,

  • Countries of different sizes
  • Asymmetric tax competition Large jurisdictions
    with some market power have tgt0, and small
    jurisdictions benefit from K flows (Wilson, 1991
    Bucovetsky, 1991)
  • New econ geography models larger core
    economies w/ agglomeration economies have higher
    tax rates, and higher G, relative to smaller,
    less developed peripheral economies
    (Baldwin-Krugman, 2000)
  • Economies of scale in G reduce underprovision as
    population increases (Wilson, 1995)

  • Different alternative taxes Mendoza-Tesar
  • Examine race to the bottom in EU in model where
    budget is balanced with (1) wage tax, or (2)
    consumption tax countries (UK/Cont Europe) are
    Nash competitors
  • Assumed labor supply elasticity is large,
    following work of Prescott (2008), and K is less
    than perfectly mobile
  • With highly distortionary wage tax, current
    equilibrium consistent w/ data no race to the
  • Get staggering race to the bottom w/
    consumption tax

  • Models with both mobile and immobile capital
  • Clear incentive for differential taxation
    (preferential regimes for mobile capital),
    including mitigation of public service
    under-provision problem
  • Implies zero tax on any perfectly mobile capital,
    or lower tax on relatively mobile capital
  • Achieve directly, or with tax preferences that
    effectively apply only to mobile capital, or with
    mix of lowering STR and raising METR (Desai,
    1999 Gugl-Zodrow, 2006 Haufler-Klemm-Schjelderup
    , 2006 DLR, 2008)

  • Growing literature on total revenue effects (but
    not welfare analyses) of preferential tax regimes
  • Preferential regimes allow higher tax rate on
    immobile capital, but lose revenues on mobile
  • Non-preferential regimes have lower tax rate on
    immobile capital, but limit tax comp over mobile
  • Quite varying results (Janeba-Peters, 1999 Keen,
    2001 Janeba-Smart, 2003 Wilson, 2006),
    depending partly on mobility assumptions,
    symmetry assumptions regarding jurisdictions,
    whether tax bases are endogenous

  • Marceau-Mongrain-Wilson (2007) allow asymmetry,
    mobile capital, and endogenous tax bases
  • If constrained to uniform taxation, some
    countries have low rates and attract mobile K,
    others have higher rates on immobile K
  • If not constrained to uniform tax, get intense
    tax comp for mobile K
  • Net result get more total tax revenue with
    non-preferential regime, even though get tax

  • Low rate countries (tax havens) are small (as
    in asymmetric tax competition models), as face
    more elastic K supply, but small in relative
    endowment of immobile K which explains data
    better than using population, where empirical
    results are mixed
  • Also, less productive countries set lower tax
    rates, and productivity differentials tend to
    make preferential regimes more attractive

  • Tax competition literature focuses on
    inefficiencies, especially downward pressure on
    public services, including social welfare
  • But some models stress positive aspects of tax
  • Limit Leviathan, special interest overspending
    on public sector (Edwards-Keen, 1996)
  • Encourage use of desirable user charges/benefit
    taxes (McLure, 1986 Huber-Runkel, 2004)

  • Limit tendency to over-tax K that is immobile in
    the short run (Janeba, 2000)
  • Limit generally undesirable taxation of capital
    income (consumption tax arguments, reductions in
    national/federation capital stocks)

  • Hong and Smart (2006) Tax competition with
    income shifting desirable because moves economy
    toward zero tax on mobile capital
  • GE model with tax planning (inter-affiliate
  • Allowing some MNC tax planning may increase home
    country welfare, as I and productivity gains
    outweigh limited revenue losses
  • FDI is less sensitive to tax rates in model
  • Potential for tax planning can raise corporate
    tax rates, since FDI costs of high rates are lower

  • Peralta, Wauthy and van Ypersele (2003)
  • Can achieve similar results by not enforcing
    transfer pricing rules
  • But, Slemrod-Wilson (2006) argue that allowing
    tax havens and tax planning reduces welfare
  • Due to costs of both planning and limiting
  • Aggravate tax competition problems
  • Especially costly since labor income taxes also
    avoided make case for some K taxation to reduce
    labor tax avoidance problems

Five Primary Qualifications
  • Clear that prediction of zero (or negative)
    source-based tax rate on internationally mobile
    capital is counterfactual, but suggests downward
  • Five primary qualifications
  • Capital taxes may not be fully shifted to labor
  • Treasury transfer arguments
  • Backstop argument
  • Implications of tax avoidance
  • Taxation of location-specific economic rents

  • 1. Tax is not all shifted to local factors
    (Gravelle and Smetters, 2006) in US context,
    labor bears gt50 in most simulations, because
  • Not small economy, so capital bears burden equal
    to fraction of world output
  • But share of world output lt 3 in Canada
  • Canadian and US capital are close substitutes
    (Cummins, 1997 Altshuler-Cummins, 1997)
  • Canadian capital markets integrated with
    international capital markets (Mintz, 2006)

  • Most models assume capital supply is highly or
    perfectly elastic (Finance Canada Cox
  • Imported goods are not perfect substitutes for
    domestic goods, so can shift capital tax to
    prices, borne in part by capital owners
  • Older import substitution elasticities estimates
    in Canada vary from low 0.1-1.5 by
    Roland-Holst, Reinert and Shiells (1994) to
    moderately high 1.0-4.8 by Cox and Harris
    (1992) and Cox (1994)

  • But trade economists argue that these are too
    low, as imply too much local market power
    (McDaniel-Ballistreri, 2002 Harberger, 2008)
  • Several recent studies get higher estimates, with
    many industries in 6-13 range (Erkel-Rouse and
    Mirza, 2002)
  • Recent estimates for Canada-United States
  • Head-Ries (2001) 7.9-11.4
  • Clausing (2001) around 10

  • And, structure of model is critical
  • Randolph extends the Gravelle-Smetters model to
    include domestic corporate sector with two goods,
    one perfect substitute with import, one imperfect
  • Import substitutability becomes far less
    important (no effect if K/L same for two goods)
  • If world output share is small, labor bears gt100
    of corporate income tax

  • Empirical evidence generally consistent w/
    significant, though not complete, shifting of CIT
    to labor
  • Devereux, et al. (2008) long run shifting to
    wages of 92 (in context of a wage bargaining
    model), using data from 9 EU countries over
  • Hassett and Mathur (2006) using data from 72
    countries over 1981-2002, find wages very
    responsive to CIT rate, but somewhat less so to
    AETR, METRs, especially in small countries
  • Desai-Foley-Hines (2007) get 45-75 shift to L in

  • On balance, suggests importance of
    Gravelle-Smetters argument limited in Canada in
    long run
  • In any case, in long run tax on capital may be
    undesirable (consumption tax arguments)
  • But imperfect mobility of capital in SR provides
    a powerful argument for capital taxation in SR

  • 2. Treasury Transfer Argument
  • US (and UK, Japan) allow FTCs for foreign taxes
    paid, up to US tax assessed on US-defined income
    (Earnings and Profits definition)
  • Suggests corporate tax at rate near US rate is
    desirable, as pure transfer from US treasury,
    with no disincentive effects on FDI
  • BUT, many qualifications to this free revenue

  • Irrelevant for territorial countries or those
    that grant tax sparing, but not important in
    Canada as US, UK and Japan are 74 of FDI
    (US64), and 82 of FDI in footloose
  • FTCs received only when funds repatriated to
  • Thus, treasury transfer effect arguably relevant
    only for FDI financed with new contributions of
  • For investment financed with retained earnings,
    foreign taxes on dividends are irrelevant
    (controversial new view of dividend taxes,
    supported by recent evidence) and only host
    country tax matters (Hartman, Sinn)

  • Many firms in excess foreign tax credit position
    so lower host country tax desirable as uses
    excess credits without incurring further domestic
  • Virtually all resource firms in Canada are in
  • gt50 of others are in E-FTC (though fluctuates)
  • AND, avoidance techniques appear to allow
    creation of E-FTCs at will, using reverse
    hybrids made possible with US check the box
    rules that allow free designation of entity as
    taxable corporation or pass through disregarded

  • Specifically, rules allow creation of hybrid
    entities (a foreign corporation but a US
    pass-through entity), and reverse hybrid entities
    (foreign pass-through entity but a US
  • US MNC can separate taxes paid to hybrid
    (foreign holding company, deemed responsible for
    all taxes paid) and deduct currently, from income
    earned by reverse hybrid (foreign operating
    company), which is deferred, perhaps indefinitely

  • AND, rules for defining income to determine FTC
    are harsh, resulting in larger deemed income
    and lower deemed taxes paid, and thus more
    taxes leading to E-FTCs and double taxation
  • Definition of income uses less generous CCAs
  • General expenses allocated to foreign firms,
    including very harsh waters edge interest
    allocation rules to be replaced by fairer
    worldwide allocation rules in 2009
  • Finance Canada estimates need rate in Canada of
    23-25, even w/ new interest rules, to avoid

  • Most empirical work suggests treasury transfer
    effects not important General tentative
  • FDI responsive primarily to host country rates,
    not home country tax rates
  • FDI financed with retained earnings is more
    sensitive to host country tax rates than if
    financed with debt or new equity transfers (new
  • Conclusion Treasury transfer effect not critical
    in designing CIT in Canada at current rates (but
    depends on US policies regarding avoidance and

  • 3. The backstop argument
  • A CIT is necessary to prevent accumulation of
    untaxed labor returns within a corporation
    (although often evaded in any case)
  • A low corporate rate, relative to personal rate,
  • Personal taxation of some returns is deferred
    until paid as dividends (fully integrated in
  • Some retained earnings taxed at personal level as
    capital gains, but others subject to exclusion
    will avoid tax

  • Shifting between bases is large in the US
    (Gordon-Slemrod, 2000)
  • Taxable income elasticity of self-employed high
    in Canada (gt1) (Sillamaa and Veall, 2001)
  • Small business sector important in Canada small
    Canadian controlled private corporations provide
    18.5 of CIT revenues
  • So want to limit differential between PIT/CIT
    rates, but optimal differential unclear

  • 4. Implications of corporate tax avoidance
  • Provides another dimension of tax competition, as
    discussed previously
  • BUT, also suggests that negative effects on FDI
    of high rates mitigated through avoidance
  • Indeed, MNC avoidance allows differentiation
    between immobile domestic firms and more mobile
    foreign firms, within a uniform corporate tax
    desirable on efficiency grounds (Desai, 1999
    Gugl-Zodrow, 2006 Hong-Smart, 2006)

  • Altshuler-Grubert (2006) argue that tax
    competition in allowing avoidance is increasingly
    important, especially income shifting to tax
  • Stress incentives for all three parties
  • MNCs lower tax burden
  • Host countries attract internationally mobile
    capital while taxing immobile domestic capital
    and capital earning location-specific rents w/
    uniform CIT
  • Home countries increase in MNC returns,
    including due to more efficient I allocation gt
    tax rev losses

  • Facilitated by US with check-the-box rules
  • Example Subsidiary in high-tax country is a
    hybrid (treated as corporation by host country
    but is pass-through entity with a tax haven
    finance subsidiary)
  • Tax haven finance subsidiary
  • gets equity from US parent
  • loans to hybrid entity in high-tax country

  • Get interest deductions in high-tax country, but
    payments to tax haven finance subsidiary do not
    trigger CFC (controlled foreign corporation)
    rules, so taxed only at tax haven rates, and not
    taxed currently in US
  • Hybrids also used to shift dividends, royalties
    without triggering CFC rules
  • So, case for low rates mitigated to extent that
    increased tax avoidance reduces negative effects
    of high tax rates on FDI

  • 5. Taxation of location-specific rents
  • Location specific rents may arise due to
  • Access to consumers, markets (including US mkt)
  • Agglomeration economies (technology, knowledge
  • Low transport costs
  • Skilled but relatively inexpensive factors
  • Productive infrastructure
  • Make higher capital tax rates desirable, as

  • Desirable to tax location-specific rents
  • Efficient source of revenue
  • Likely to be important due to access to US market
  • Especially desirable politically if earned by
    foreigners, as is increasingly likely as
    globalization advances (Huizinga-Nielsen, 1997)
  • Countries with large foreign ownership share have
    higher corporate tax burdens (Huizinga-Nicodeme,

  • Can some rents be taxed with alternative
    instrument, reducing argument for taxation with
  • Resource rents federal production taxes,
    sector-specific cash flow tax?
  • Rents in protected financial sector cash flow
    tax on net interest (report of Presidents panel
    in US)
  • But rents due to access to US market presumably
    must be taxed with CIT, providing strong argument
    for taxation of capital income

More Empirical FDI Tax Sensitive?
  • Perfectly elastic supply not likely
  • But, much evidence of significant elasticity,
    increasing over time
  • Gordon and Hines (2002) econometric work of the
    last fifteen years provides ample evidence of the
    sensitivity of the level and location of FDI to
    its tax treatment
  • Similar conclusions by de Mooij and Ederveen
    (2003, 2005)

  • De Mooij and Ederveen (2003) calculate median
    investment tax elasticity of 3.3 nearly double
    that for new plants most recent studies have
    largest elasticities (although considerable
  • Altshuler, Grubert and Newlon (2001) estimate
    investment tax elasticity has increased from 1.5
    in 1984 to 2.8 in 1992
  • Altshuler and Grubert (2004) get estimates near 4
    for 2000

  • In Canada,
  • Iowerth and Danforth (2004) estimate a combined
    domestic and foreign investment elasticity of
    approximately 1
  • Department of Finance Canada (2007) estimates a a
    combined domestic and foreign elasticity of 0.7
  • In US, Hassett and Newmark (2008) confirm
    sensitivity of investment to tax factors

  • Devereux and Griffith (2003) find significant tax
    effects of AETR in attracting investments that
    generate above-normal returns (firm-specific
  • Measured AETR is weighted (by above-normal and
    normal returns) average of STR and METR
  • STR key in choice among competing locations
  • METR key in level of investment, given location

Is There Evidence of Tax Competition?
  • Various attempts to measure tax competition
  • Changes in corporate statutory tax rates
  • Average statutory tax rates in OECD dropped from
    about 40 (1965-late 1980s) to 32 in 2004
    (Devereux, 2007) roughly constant since 1993
  • Average statutory tax rates in EU and U.S. fell
    even more dramatically, e.g., 48 in 1982 to 35
    in 2001

  • In early 1980s, CIT statutory tax rate in OECD
    was 50 greater than labor tax wedge, but by 2000
    roughly equal, with larger drops in smaller
    countries (Haufler, Klemm, Schjelderup, 2006)
  • Reduction in combined federal provincial
    statutory corporate income tax rate in Canada
    from 45 in 1980 to 35.0 in 2006 to 27.6 in
    2012, with goal of 25
  • Similarly, average tax rates on highly profitable
    I declined

  • But marginal effective tax rates declined
    relatively little, much less than might have, due
    to base-broadening reforms (dispersion of
    allowances down)
  • Corporate tax revenues/GDP roughly constant for
    40 years and indeed increased slightly (Devereux,
    2007 Stewart and Webb, 2006)
  • Corporate income tax share of tax revenues also
    roughly constant in OECD in US and Canada,
    increased over the past 20 years (Auerbach, 2006)

  • So, tax comp as measured by rates limited
    primarily to reductions in STRs in OECD (income
    shifting tax comp)
  • Not yet a race to the bottom among OECD countries
  • But average foreign corporate tax burdens on US
    foreign affiliates have declined markedly 43
    in 1982 to 26 in 1999 (Hines, 2008)
  • And much smaller differences in STRs, AETRs
    across large and small countries, by population,
    suggesting more tax competition over time for
    larger countries (Hines, 2008)

  • Note Many potential reasons for constant CIT/GDP
  • Increased relative profitability
  • Increased investment with lower rates
  • Increased above-normal returns for profitable
  • Increased income shifting from individual income
  • Increased profit shifting as rates decline (if
    average CIT/GDP is unweighted, increased by
    income shifting to smaller lower-rate countries)

  • Increased openness results in lower corporate
    income tax rates, tax burdens
  • Winner (2005)
  • Measures capital mobility across sample of 23
    OECD countries, using S-I correlations
  • Finds decreasing capital tax burdens and
    increasing labor tax burdens, increasing over
    time, especially for small countries
  • Similar results on CIT rates Bretschger-Hettich

  • Haufler, Klemm and Schjelderup (2006)
  • Larger share of mobile capital (modeled as share
    of value added in manufacturing relative to
    services) has significant negative impact on
    revenue mix (corporate revenues/wage taxes)
  • Slemrod (2004)
  • STRs decrease with measure of openness
  • But, not CIT revenues/GDP

  • Most direct evidence of tax comp
    (Devereux-Lockwood-Redoano, 2008) among (10)
    OECD nations, find tax comp in STRs defined as
    reduction in STR in response to reduction in
    weighted average STR of competitors
  • Find weaker evidence for tax competition in METRs
  • Argue that tax competition in STR is dominant
    form trying to attract mobile firms with
    firm-specific rents, while maintaining revenues
  • Also find that the empirical pattern of recent
    STR reductions consistent with their theory

  • Tax competition also tested directly by
    Heinemann, Overesch and Rincke (2008)
  • Examine discrete tax reform events statutory
    rate-cutting reforms (sample of 32 European
  • Also get significant reforms involving reductions
    in own STRs likely in response to STRs of
    neighboring countries, relative to own rate (cut
    rates by 1.5-3.2 points if competitors cut by 1
  • Rate cuts more likely during election campaigns

  • Broadly similar results
  • Altshuler and Goodspeed (2002) EU has a
    positive tax reaction function, with US a
    Stackelberg leader in tax rates after 1986, and
    EU-US tax competition increasingly intense
  • Besley-Griffith-Klemm (2001) CIT capital tax
    rates positively related to same tax rates in
    other countries, especially in EU, but not for
    other taxes
  • Many similar results for states/provinces, local

  • Evidence of more rapidly declining ATRs in
    smaller less developed countries
  • Sample of 60 countries indicates decline of 10
    percentage points between 1984-1992, with larger
    declines in small, open and relatively poor
    economies (Grubert, 2001)
  • Decline continued but moderated through 1997, and
    stopped 1998-2000 (Altshuler and Grubert, 2006)

  • From early 1990s to early 2000s, revenues and
    AETRs have declined in developing countries
    primarily due to tax holidays, tax incentives for
    FDI (Keen-Simone, 2004)
  • Average STRs declined by 6 percentage points
  • CIT Revenues/GDP declined by 20
  • Significant expansion of tax holidays, tax
    incentives, especially in poorest developing

  • Similarly, Huizinga-Nicodeme (2006) find CIT/GDP
    declines with size, if size is measured as GDP,
    and tax rate reductions yield revenue declines in
    new members of EU
  • Suggests more of both STR and AETR/METR tax
    competition in developing countries
  • But questioned by Hines (2006), who argues that
    for longer time period and different sample of
    non-OECD countries that CIT/GDP and CIT/Total TR
    roughly constant over 1972-2004 (but not if start
    in early 1980s, especially for former)

  • Similarly, Garretson and Peters (2006) argue tax
    competition varies across OECD countries
  • Increased capital mobility, defined as larger
    fraction of FDI/I or capital mobility
  • Find evidence of tax competition in statutory
  • But much less intense for core countries with
    agglomeration economies (large market potential)
  • Canada low to moderate tax competition because
    3rd or 9th of sample of 19 in agglomeration
    economies (depending on measure)

  • Altshuler-Grubert (2006) measure of tax comp
  • Find that ATRs on US multinationals decrease
    significantly if a country is losing share of US
    foreign capital stock to others
  • Especially true if small, by population
  • (But relationship stops after 1997)

Alternative Explanations
  • STR reductions reflect not tax competition, but
    common intellectual trends, especially for
    classic base-broadening, rate-lowering reform,
    following UK and US examples in mid-1980s
  • STR reductions reflect not tax competition, but
    yardstick competition where lower rates mimic
    low rates of nearby countries used as benchmarks
    by residents

  • Devereux, Lockwood and Redoano (2008) test these
  • Analyze level of capital controls in taxing
    country and its neighbors
  • If tax competition is common intellectual trends
    or yardstick, argue capital controls should be
  • Instead, they find tax competition exists only
    between countries without capital controls

  • Altshuler-Grubert (2006) Empirical evidence that
    tax competition as tax avoidance more important
  • Correlation between statutory and effective tax
    rates declined significantly
  • Reported profitability of tax havens increased
  • Intercompany transfers increased considerably
  • In 2002, estimate US MNCs saved 15 of foreign
    tax burden (7 billion)
  • But no international evidence yet that
    sensitivity of FDI to host country tax rates is
    declining over time

  • Should tax competition be a significant factor in
    determining corporate income tax policy,
    including in Canada?
  • Certainly, to some extent
  • Very difficult to test for tax competition, BUT
  • Much evidence of tax sensitivity of FDI
    essential assumption of tax competition for real
  • Much evidence of income shifting, and greatly
    facilitated by US policy choices

  • Much evidence of tax competition in form of
    statutory tax rates (both STR patterns and tax
    reaction functions), and ATRs, especially
    effective tax rates on highly profitable
  • But little on the METRs, that are a key factor of
    tax competition for real investment
  • Suggests tax competition most fierce for paper
    profits, and firm-specific economic rents that
    generate important externalities

  • A plausible explanation of events thus far
    i.e., find ways to lower tax mobile capital,
    while maintaining revenues from taxing relatively
    immobile capital
  • Analogous to tax competition in US states state
    and local government provide multitude of
    individualized tax incentives to highly mobile
    capital (Zodrow, 2003)

  • BUT, to explain STR declines, METR constancy
  • Common intellectual trends argument likely to
    have some validity
  • Most of STR reductions in 1980s
  • Base-broadening rate-reducing reforms driven
    primarily by domestic concerns
  • Plus, treasury transfer effect pressures on rates
  • In US, revenue neutrality requirement increased
    attractiveness of base broadening with rate
    reduction, coupled with many standard arguments

  • Arguments for lower rates not affected much by
    Gravelle-Smetters arguments, treasury transfer
    effect or, to a smaller extent, backstop PIT
  • But significantly tempered by benefits of taxing
    location specific rents, esp. w/o alternative
  • And perhaps by role of avoidance techniques

  • Role of tax avoidance is complex
  • Encourages tax competition in STR
  • Reduces importance of treasury transfer argument
    for higher rates
  • BUT, mitigates costs of high rates find
    evidence on declining sensitivity of investment
    with avoidance?
  • And US rules could change, especially if new
    Democratic administration

  • Also, constancy of CIT/GDP reflects many non-tax
    factors that may change (esp. high profitability,
    more incorporation)
  • Keen Tax competition in STRs shows no signs of
    abating with Canada a recent example
  • But scope of base-broadening with further STR
    reductions is limited likely to affect METRs,
    ATRs much more if trend continues
  • And revenues, especially if profitability declines

  • Recently enacted Canadian reforms seem a good
    response to current environment
  • STRs and METRs are lowest of G-7
  • Comparable to smaller developed countries and
    emerging/developing countries (with lower METRs
    but other disadvantages) that are likely to
    become more significant competitors for mobile
  • BUT, tax competition likely to continue,
    especially if US lowers CIT rate, as increasingly

  • Will result in tougher future choices
  • Decline in STR will reduce taxation of rents
    (might be mitigated with alternative tax
    instruments) and AETRs and METRs, as CCAs are
    generally appropriate
  • Competition with smaller developing countries and
    emerging nations may become more intense with
    more globalization, lower transport costs,
  • Further reductions may be appropriate by 2012

  • Lower tax rates at corporate level may be
    accompanied by higher tax rates at personal
    level, on labor or capital income, as former
    determine both real FDI and income shifting
  • Or may consider more drastic reforms
    consumption-based taxes or dual income taxes
  • Key question Will competition in STRs in fact
    continue, given fewer opportunities for
  • Key research areas Models that simultaneously
    include all aspects of various forms of tax
    competition better understanding of impact of
    developing, emerging nations

The End
Last Revised
  • June 6, 2008