CHAPTER 2

Why and How to Tax

Corporate Income
Ruud de Mooij and Alexander Klemm

INTRODUCTION
The question whether countries should tax corporations on their income is an old
one, yet it has never been more topical than it is today. Tax competition is driving
down corporate tax rates in a race to the bottom, and 12 countries today levy no
corporate income tax at all. The pressure from tax competition and the major
complications that arise in administrating and enforcing current corporate
income tax systems are so vast that one may wonder whether it is worth dealing
with them or whether it would not be much easier simply to give up on taxing
corporate income, replacing lost revenues with other taxes.
This chapter starts by discussing why it is worthwhile for countries to main-
tain a corporate income tax, even if possibly in a reformed manner. It starts, in
the first section, by discussing the role of the corporate income tax as part of the
broader income tax system. Corporate income is a subset of capital income, and
so the first question to answer is whether capital income should be taxed or not
(“Why Tax Capital Income?”). If that is answered in the affirmative, the next
question is whether there is a role for the corporate income tax in the enforce-
ment of capital income taxation (“Why Tax Corporate Income?”). If one accepts
that, the subsequent issue that arises is how corporate income should be taxed
and how such a tax should be implemented. While details of corporate income
tax systems around the world differ, there are some common principles underly-
ing them—including with respect to the taxation of international business
income (“The Standard Corporate Income Tax”). Yet there are alternative sys-
tems that are based on different principles and that might avoid some of the
distortions and complexities of the current corporate income tax. The chapter
discusses in particular the comprehensive business income tax, which broadens
the base of the corporate income tax by disallowing deductions for interest
expenses (“Comprehensive Business Income Tax”). Finally, the chapter discusses
various implementations of rent taxes, which exempt the normal return on cap-
ital at the corporate level.

11

©International Monetary Fund. Not for Redistribution

12 Corporate Income Taxes under Pressure

WHY TAX CAPITAL INCOME?

Income generally arises from labor effort, be it wages or entrepreneurial activity,
and capital returns in the form of interest, dividends, or capital gains. The dis-
tinction between labor and capital income is not always easy to make. For
instance, self-employed entrepreneurs do not distinguish between payment for
the labor effort they put into their own company and their capital investment.
The entrepreneurial income they earn is simply the sum of both. Attempts to
distinguish labor and capital income for tax purposes then become arbitrary and
prone to avoidance: entrepreneurs can easily present income as either labor or
capital income, depending on which is being taxed least.
This arbitrage between capital and labor income is at the heart of discussions on
the design of the income tax system. For some, the theoretical ideal is to tax the sum
of labor and capital income at a progressive rate structure, consistent with the ability-
to-pay principle. This “global income tax” prevents arbitrage between labor and
capital that could otherwise arise for the taxation of self-employed entrepreneurs.
Others have argued, however, that only labor income should be taxed, while
capital income should be exempt. Their argument goes as follows: a tax on (the
normal return to) saving increases the relative price of future consumption rela-
tive to present consumption. Chamley (1986) and Judd (1985) show that, in an
infinite horizon model, this is inefficient and violates horizontal equity principles:
it is always better to tax only labor income and avoid the intertemporal distortion
to savings. The optimal tax on the normal return is therefore zero (see also
Atkinson and Stiglitz 1976). This result has led economists to argue for an
exemption of the normal return (see, for example, Mirrlees and others 2011).
However, the income tax literature has also criticized the Chamley–Judd outcome
for relying on too-specific assumptions, such as optimization over an infinite
horizon. Models that relax the assumptions arrive at different conclusions and
offer several rationales for a positive capital income tax rate, both for efficiency
reasons and to address equity concerns (see, for example, Banks and Diamond
2010). Recent contributions further suggest that the classical Chamley-Judd
results are misinterpreted, providing further theoretical ground for the taxation of
capital income (Straub and Werning 2020). Overall, the theoretical consensus has
been shifting away from the question whether capital income should or should
not be taxed towards the question to what extent it ought to be taxed.
Yet others favor an intermediate position between comprehensive taxation of
income and zero taxation of capital income. They support separate taxation of
labor and capital income under a “dual income tax,” with a progressive rate
scheme applying to labor income, and with a flat rate applying to capital, usually
at a relatively low rate. The motivation for this is twofold: first, the lower tax on
capital mitigates distortions in saving and investment, which tend to be relatively
severe; second, flat capital income taxes do not need to be personalized, which
eases enforcement by using withholding schemes.
When discussing the taxation of capital income, it is useful to distinguish
between the normal return on capital and economic rents. The normal return on
capital is generally defined as the minimum return required to make investors

©International Monetary Fund. Not for Redistribution

 Chapter 2 Why and How to Tax Corporate Income 13

equally well off (with an adjustment for risk) compared to some benchmark
investment, such as a government bond. The remaining profit, over and above the
normal return, is called “rents.” While the normal return can be called capital
income, rents might in fact be subject to bargaining between workers and capital
owners—and thus can be reflected either as capital income or as labor income (in
the form of higher wages).
Public finance literature is unanimous in advocating taxes on rents as they are
in principle nondistortionary. A classic result from theory is therefore that they
should in fact be taxed at 100 percent. Practical considerations lead to lower tax
burdens, however, because rents are also often “quasi rents,” arising from specific
investments with a fixed cost. Rent taxes are also deemed equitable since assets are
typically owned by the well-off. The more controversial debate outlined earlier in
this section on whether capital income should be taxed thus refers to the normal
return, not to economic rents.
Another important dimension of capital income taxes arises in open econo-
mies, as the supplier of the capital might be based in a different country from
where capital is demanded and used. Here, there are two principles of taxation:
source or residence. “Source” refers to the country where the capital is installed
and where it yields its return. “Residence” is the country where the owner of the
capital resides (see Chapters 3 and 4). Residence-based capital income taxes are
consistent with taxation based on the ability to pay of domestic residents, as they
are taxed on their worldwide capital income—irrespective of where the source of
that return is. This generally applies to the personal income tax. Source-based
taxes impose tax on investment returns irrespective of their residence. This gener-
ally applies to the corporate income tax, which thus imposes tax on the return
earned by foreign investors. Again, the distinction between normal returns and
economic rents is important. For instance, if after-tax normal returns on invest-
ment are fixed on world capital markets, any source-based tax on the normal
return will lead to adjustment in the amount of capital such that the before-tax
return rises enough to restore equilibrium. Less capital means lower wages, and so
the incidence of a source-based tax on the normal return will fall on workers.
This, however, will not apply to location-specific rents: to the extent that a
source-based tax reduces this rent, the foreign owner will not be able to escape the
tax and will bear its full incidence.

WHY TAX CORPORATE INCOME?

There are different types of systems to tax capital income (detailed in the follow-
ing paragraphs). The so-called classical corporate income tax considers corpora-
tions as separate entities from their ultimate owners. As wages and interest are
generally deductible, the corporate income tax effectively becomes a withholding
tax on equity returns at the company level. Corporate income tax is thus levied
separately, besides the personal income tax on equity returns (in the form of div-
idends and capital gains). Hence, the classical corporate income tax system
implies double taxation. If the combined burden of corporate income tax and

©International Monetary Fund. Not for Redistribution

14 Corporate Income Taxes under Pressure

personal income tax is higher for corporate businesses than for sole proprietor-
ships, it could induce entrepreneurs to run their businesses in the latter legal form
so as to avoid paying the double tax. To mitigate such distortions from the
double-tax burden on equity returns, classical systems often apply relatively low
flat personal income tax rates on equity income.
In contrast to the classical corporate income tax system, the so-called integra-
tion system looks through the corporation and acknowledges that, ultimately,
legal entities are owned by people. As only people can ultimately bear the true
incidence of taxes, this clearly makes more economic sense. The corporate income
tax then operates as a withholding mechanism for the full income tax, and impu-
tation credits are provided for individuals when calculating their personal income
tax liability. The difficulty with integration systems, however, arises from interna-
tional transactions: imputation credits are generally not provided for foreign
corporate income tax paid. This can lead to distortions in international capital
markets. In the European Union, imputation systems have therefore been abol-
ished, as they are an infringement to the freedom of establishment.
As the integration system illustrates, governments do not necessarily have to
tax capital income through corporations. Indeed, the alternative would be to tax
capital income solely and directly on the income that people receive, that is,
through personal income tax, without any withholding of tax at the corporate
level. So why is there a corporate income tax? There are two main reasons—none
of which is of a fundamental nature; they rather provide a pragmatic case for its
existence.1
The first reason is based on the relative ease of administration and particularly
the withholding role of the corporate income tax. Corporations are convenient
collection agents for governments. For instance, besides corporate income tax,
value-added tax (VAT) and personal income tax (through pay-as-you-earn
schemes) are also often withheld by corporations, as they hold proper books and
records and can efficiently be monitored by tax inspectors. Relying on individuals
or consumers to pay their tax, based on filed tax returns, would be considerably
costlier to enforce. The withholding role is especially important for profits that are
retained in the company, as distributed profits can also be taxed through a with-
holding tax on dividends. Retained profits, however, simply lead to higher value of
the firm and, therefore, to capital gains for the owners. For practical reasons, cap-
ital gains at the personal level are seldom taxed on an accrual basis but rather upon
realization. Capital owners can then postpone their tax payment by not realizing
these gains. Moreover, capital gains in small nontraded companies are especially
hard to value and might escape taxation altogether. The attraction of the corporate
income tax is that it withholds tax on all profits as they arise, thus eliminating the
difficulty in taxing capital gains and the distortions it would induce.

1
Other reasons that are often mentioned are less convincing. These include (1) the benefit principle,
that is, the idea that corporates should pay something for the use of public goods, and (2) the value of
limited liability. The weakness of these arguments is that the amount of tax paid has little relationship
to these benefits.

©International Monetary Fund. Not for Redistribution

 Chapter 2 Why and How to Tax Corporate Income 15

The second reason for taxing corporations relates to taxing the rents earned by
international businesses. Under the current international tax convention, source
countries have the primary taxing right to tax multinational income. To the extent
that this reflects a normal return, it might be shifted onto employees—as was
noted before. But to the extent that it reflects economic rent, it would fall on
foreign capital owners. For an individual country, taxing foreign capital owners on
their rents has significant appeal and provides a major attraction to the corporate
income tax. The taxation of rents that accrue to foreigners is particularly important
if rents arise from natural resources. Numerous developing countries have abun-
dant resources that are extracted by multinational companies that have their resi-
dence elsewhere. Fiscal regimes are generally in place to ensure that a significant
portion of these natural resource rents accrues to the local governments.

THE STANDARD CORPORATE INCOME TAX

The most common corporate income tax system applies tax on corporate profits,
defined along similar principles as accounting profits but with some adjustments.
The tax definition may, for example, be more prescriptive to reduce the amount
of judgment in calculating tax bases2 and to simplify compliance and enforce-
ment. Deviations could also occur for policy reasons, as governments can encour-
age or discourage certain behaviors by changing their tax implications.
Using a definition of profits as the tax base has the implication that, as in
accounting, investment is not a deductible expense. As the company merely
changes one type of asset (cash) for another (capital), such a transaction is not a
cost. The cost to the company is, instead, the loss of value of the capital due to
obsolescence or wear and tear, and this depreciation is deductible. The permissi-
ble deduction for depreciation is in many countries specified by law, for example
as an annual percentage of the acquisition cost (“straight line”) or the written-
down value (“declining balance”). Such rules are meant to prevent abuse and to
reduce compliance and administrative costs, but they can also be used to purpose-
fully encourage investment. Hence, the exact depreciation allowance for tax
purposes is likely to differ from true economic depreciation, that is, the exact loss
in value of the asset, and it may also differ from accounting depreciation,
although in some countries both are aligned.
Also, following the accounting logic, issuing or repaying debt is treated as a
change in the asset composition with no impact on profit, while interest is treated as
a deductible expense. Profits are taxed, irrespective of whether they are retained in
the company or distributed as dividends. The result is a difference in treatment in
interest and dividends, only the former of which is deductible. This generally implies

2
Judgment is often allowed in accounting, and, combined with explanatory notes, can improve the
quality of information. It would not be good to allow firms judgment in determining their tax base;
moreover, there is a strong benefit to the quality of accounting if judgmental decisions can be made
without impact on taxation.

©International Monetary Fund. Not for Redistribution

16 Corporate Income Taxes under Pressure

a debt bias in corporate income tax systems, although this also depends on the inte-
gration with personal-level income taxes: if personal income tax on interest is higher
than on dividends, this may offset the discriminatory effect associated with the cor-
porate income tax. It is unlikely to completely undo it, though, because there are
typically some taxpayers (for example, pension funds, foreign investors) who benefit
from a tax exemption or rate reduction on both interest and dividends.
Debt bias is not only a distortion that exists in theory but has been shown to
be of significant importance empirically. A meta-analysis by Feld, Heckemeyer,
and Overesch (2013) finds that the debt ratio of nonfinancial firms increases on
average by 0.27 percentage point for each percentage point increase in the corpo-
rate income tax rate. Another metastudy by De Mooij (2011) finds that these
responses are increasing over time, suggesting that distortions relating to debt bias
have become more important. De Mooij and Keen (2016) and Luca and Tieman
(2016) analyze debt bias among, respectively, banks and nonbank financial firms.
Both find significant and sizable effects on leverage ratios, which is particularly
worrisome in light of the implications high debt can have on financial stability.
The IMF (2009; 2016) argues that such biases are hard to justify, and alternative
systems have been proposed by economists that address this bias.
Standard corporate income tax systems not only distort the financial structure of
corporations, but also the level of investment. By allowing a deduction for both
depreciation and interest, marginal investments that just break even and that are
financed by equity are discouraged by taxation as taxes increase the cost of capital.

COMPREHENSIVE BUSINESS INCOME TAX

If interest deductibility creates debt bias, the obvious solution is not to allow it.
A US Treasury report (Department of the Treasury 1992) works out a concrete
proposal for how such a system would work, including its international implica-
tions. Clearly, to prevent double taxation, such a proposal also means that interest
receipts of corporations are untaxed. This would also apply to banks and other
financial intermediaries, which would no longer be taxed on profit associated
with the interest margin between lending and borrowing. Moreover, the personal
income tax on interest might be abolished or reduced to bring it in line with
dividends and capital gains. The main impact of a comprehensive business
income tax is hence to move the point of ultimate taxation from the lender to the
borrower. So instead of collecting corporate income tax from financial intermedi-
aries and personal income tax on interest from any savers providing funds, the
comprehensive business income tax would collect the tax in one step from the
borrower.3 This would thus mirror the treatment of dividends, which are equally

3
The specific comprehensive business income tax proposal by the US Treasury argues for exempting
interest and dividend income throughout, but, in principle, additional personal-level taxation could
be maintained, as long as treatment on interest and dividend is the same. Such treatment would be
required if progressive taxation of comprehensive income is desired. A governmental commission in
Sweden proposed the introduction of a comprehensive business income tax in 2014.

©International Monetary Fund. Not for Redistribution

 Chapter 2 Why and How to Tax Corporate Income 17

nondeductible, and—at least under certain circumstances—typically nontaxable

when received by corporations.
A comprehensive business income tax would effectively mean that the normal
return on capital, irrespective of whether financed by debt or equity, is taxed in
the source country and at the level of the firm. This would raise taxes and redis-
tribute revenues across countries. The increase in taxation would be the result of
current tax structures, under which some interest recipients are exempt from tax
or are tax favored. Pension funds, for example, are often tax exempt, implying that
their interest income is not taxable. A shift to a comprehensive business income
tax would change this, as the pension funds’ interest income would be taxed at
the corporate level of the firms in which they invest. International redistribution
would occur because foreign debtors are currently only subject to a possible with-
holding tax on interest in the country where they invest. These tax rates are typ-
ically lower than full corporate income tax or personal income tax rates and may
even be zero in some cases (notably if the investor is a sovereign wealth fund). The
country where the investor is located may, in turn, collect taxes, typically credit-
ing the withholding tax. A move to a comprehensive business income tax would
mean higher revenues for the source country, and either lower revenues in the
residence country (if the system is adapted there too) or higher total taxation of
interest (assuming no credit is given for corporate income tax).
A comprehensive business income tax effectively taxes the return on debt at
the level of the corporation, instead of at the level of the owner of the capital. If
this is to be integrated with the personal income tax, it serves as a withholding tax
for the personal income tax. Yet a comprehensive business income tax can also
serve as a final withholding tax on all capital returns if the personal income tax
on interest, dividends, and capital gains is abolished. By taxing the normal return,
a comprehensive business income tax would discourage investment. The inci-
dence of the comprehensive business income tax might thus to a significant extent
be shifted onto workers in the form of lower wages.

RENT TAXES
Instead of taxing the full normal return to capital at the corporate level, as under a
comprehensive business income tax, another class of corporate tax systems exempts
the normal return entirely at the corporate level. These can be labeled “rent taxes.”
There are different ways to implement rent taxes, as this section will discuss. Note
that normal returns can still be taxed at the personal level. The systems discussed
here refer only to the taxation of capital returns at the corporate level.

Cash-Flow Taxes
One class of rent taxes is known as cash-flow taxes, first systematically classified
in Meade (1978). The simplest form is the real (“R”)-base cash-flow tax, first
described by Brown (1948). As suggested by its name, it is defined as the net sum
of all real—meaning nonfinancial—flows. On the incoming side, this is simply a
company’s sales, including of capital goods. On the outgoing side, this includes

©International Monetary Fund. Not for Redistribution

18 Corporate Income Taxes under Pressure

all costs, such as labor and purchases of intermediate and capital goods. Financial
flows, such as interest payments, net debt issuance, and net dividends, are exclud-
ed from the tax base.
Compared to the standard corporate income tax, there are two key differences:
First, disregarding financial flows means that interest is nondeductible. This
aspect of the cash-flow tax is similar to a comprehensive business income tax, and
it means that there is no debt bias in the system, as dividends also remain nonde-
ductible. Second, investment can be immediately expensed. This deduction of the
cost of capital goods is significantly more generous than the depreciation allow-
ances under the standard corporate income tax system: while such allowances also
permit deduction of the full cost over the lifetime of a capital good, the amount
is worth less in net present value terms (and if, as is typically the case, there is no
adjustment for inflation, there is a real erosion of the cumulative value of deduc-
tions). This more generous treatment of investment implies an exemption of the
normal return on assets from tax, restricting the tax base to economic rent (see
Box 2.1). As a result, the cash-flow tax is neutral with respect to investment: any
investment that is worthwhile in the absence of taxation remains profitable under
such tax, because the investors are not taxed on their required rate of return.
Various tax reform proposals incorporating R-base cash-flow taxes have been
made. For example, the Hall and Rabushka (2007) flat tax is essentially an R-base
cash-flow tax combined with a flat-rate tax on wages. Bradford’s (2004) X tax is
equally an R-base cash-flow tax, but in this case combined with a graduated tax
on wages. Moreover, Bradford discusses a destination-based version of this tax, a
topic that is covered in Chapter 13.
A second type of cash-flow tax is levied on real and financial (“R+F”) cash-
flows. The inclusion of financial flows means that interest deductibility is back, but
changes in net debt are also included. Specifically, any inflows from issuing loans
are taxed, while outflows, such as repayments, are deductible. This ensures the
neutrality of the tax: if interest deductibility were simply combined with expensing
of investment, the result would be subsidized investment, as the expensing alone
already achieves neutrality. By taxing the debt issuance, this effect is undone.
The R and R+F cash-flow taxes thus both tax economic rents. The difference
is mostly in implementation and in where tax is collected, but there is one excep-
tion. Suppose the financial sector earns economic rents in its lending activity.
Under an R+F base, such rents are always taxed. Under an R-base tax, the taxation
of rents depends on the type of borrower. If the borrower is a taxable business,
then rents are taxed because interest is not deductible. If the borrower is, however,
an individual not covered by the cash-flow tax, then rents would go untaxed,
unless another mechanism to recover tax on them is introduced (for example, if
interest is not deductible for individuals, it is covered by the personal income tax).
A third type of cash-flow tax discussed by Meade (1978) is the share transactions
(“S”)-base cash-flow tax. This simply taxes the net distributions of companies; that
is, dividends and share buybacks are taxable, while capital increases are deductible.
In net present value terms, this tax can be shown to be identical to an R+F-base tax.
It is therefore also neutral. To see this intuitively, note that capital raised is deduct-
ible, which results in the same effect as the deduction of investment in Box 2.1.

©International Monetary Fund. Not for Redistribution

 Chapter 2 Why and How to Tax Corporate Income 19

Box 2.1. Why Does Expensing Capital Imply Rent Taxes?

The R-base cash-flow tax, which allows expensing of investment but no deduction for
interest, is an example of a tax on economic rents. As it may not be intuitive why expensing
implies rent taxation, the following provides an illustration.
Suppose a firm plans an investment I held for one year, yielding a rate of return p, and
facing a cost of capital r. In the absence of taxes, the economic rent of such an investment
would be:
I (1+ p ) p−r (1)
R = −I + =I .
1+ r 1+ r
A cash-flow tax imposed at rate t would then raise the following present value tax
payment T:
tI (1+ p ) p−r (2)
T = −tI + = tI .
1+ r 1+ r

This equation shows that no tax is payable in present value terms whenever the rate of
return equals the cost of capital at which the firm discounts future flows. If the govern-
ment’s discount rate is lower, the present value of tax revenue would still be positive.
Rent-earning investments would be taxed, but tax would never turn a profitable invest-
ment into a loss-making one. The cash-flow tax is therefore a neutral tax. This can also be
seen by deducting tax (equation (2)) from pretax rents (equation (1)), which yields R(1 – t).4
Another interpretation is that the government turns into a silent partner of the investment,
cushioning the firm’s losses in case of a bad outcome and receiving a share of the upside.
To achieve neutrality in practice, tax authorities need to refund (or carry forward with inter-
est) negative tax liabilities when investment is undertaken (left part of equation (2)).
In a traditional tax system, investment is not deductible, so the tax payment would be:
tIp (3)
T= .
1+ r

The rent net of tax would then be:

p (1 − t ) − r
R=I . (4)
1+ r

Hence, a project that just breaks even before tax (p = r) turns unprofitable. To be under-
taken, an investment needs to yield a higher return, specifically, the cost of capital is
increased from r to r/(1 – t).

4
These results continue to hold if the asset is held forever or is subject to depreciation d. In that case, the
∞
p (1 − d ) I p−d−r
pretax rent would be R = − I + ∑ =I . The cash-flow tax allows deduction of investment.
i =1 (1 + r )
i
r+d
Depreciation is not deductible, so that the posttax rent is still R(1 – t). With or without tax, the investment is
worthwhile if the real return (p – d) exceeds the cost of capital r.

Allowances for Corporate Equity or Capital

A different class of rent taxes achieves the exemption of the normal rate of return
by offering directly an allowance for it (see Boadway and Bruce 1984). One way to
implement this in a way that is relatively close to the current system is the allowance
for corporate equity. This maintains deductibility of interest on debt and

©International Monetary Fund. Not for Redistribution

20 Corporate Income Taxes under Pressure

complements it with a deduction for notional interest on equity, where the notional
interest rate is fixed administratively. This method achieves approximate rent taxa-
tion, and works better the more closely the prescribed notional interest rate matches
the normal rate of return. It also reduces the debt bias but may not fully eliminate
it if the interest rate on debt exceeds the notional interest rate on equity.
A related further step is the allowance for corporate capital, which is similar to
an allowance for corporate equity, but replaces the deductibility of actual interest
on debt with the same notional interest rate as applied on equity. This always
achieves a full abolition of debt bias at the corporate level, and achieves rent tax-
ation, whenever the notional rate is set at the level of the normal rate of return.
The allowance for corporate equity and the allowance for corporate capital are
consistent with rent taxation (assuming an appropriate notional rate is chosen)
without a need for expensing. Hence, the current depreciation system can be kept
in place. Interestingly, one of the features of these taxes is that the depreciation
rate becomes irrelevant for tax purposes. A higher depreciation rate provides a
greater initial deduction but also reduces a firm’s equity and hence the amount on
which the notional interest is applicable. It therefore even allows expensing, in
which case investment does not add to equity at all, and the allowance for corpo-
rate equity or allowance for corporate capital mirrors a cash-flow tax.

Practical Experiences with Rent Taxes

In practice, only the allowance for corporate equity has been implemented, some-
times only for a few years, in a number of countries (for example, Belgium,
Croatia, Cyprus, Italy, Latvia, Liechtenstein, Malta, Turkey). Other countries
moved partially toward an allowance for corporate equity by allowing a reduced
tax rate on the notional return (Austria), by offering it to small firms (Portugal),
or by restricting it to dividends paid out (Brazil).5
Pure cash-flow taxes are rare, but there are many examples of countries imple-
menting some of their features. Many countries, for example, have temporarily
allowed expensing of investment, but without restricting interest deductibility.
This leads to subsidization of marginal investments rather than neutrality. Some
countries use cash-flow tax features on surtaxes, for example in the natural
resource sector, to capture resource rents. Mexico had a tax that came close to an
R-base cash-flow tax,6 but it served as a minimum rather than final corporate tax.
A few countries have taxes that are charged only on distributions, thus resembling
S-base cash-flow taxes (for example, Estonia), but they do not offer a deduction
for capital increases.
Finally, the VAT also allows expensing of investment and denies interest
deductibility. It thus resembles an R-base cash-flow tax, with the major difference

5
For details of the years these taxes were applicable, as well as their main features, see Table 2 in
Hebous and Klemm (2020).
6
Impuesto Empresarial a Tasa Única (IETU, “Flat Rate Business Tax”), applicable from 2008 to
2013.

©International Monetary Fund. Not for Redistribution

 Chapter 2 Why and How to Tax Corporate Income 21

being the nondeductibility of labor costs. Value added is the sum of profits and
wages, but the VAT is actually a tax only on rents and wages, leaving normal
profits untaxed.

CONCLUSION
While taxing corporate income is common around the world, the case for doing
so is certainly not undisputed. First, the role of the corporate income tax as a
withholding mechanism at the source of profits has come under intense pressure
due to profit shifting, distortionary effects on (international) investment and
corporate finance decisions, and ongoing tax competition. Second, economists
have raised serious doubts as to the fundamental question of whether the normal
return to capital income should be taxed at all, in light of the distortionary effects
these taxes have on saving and investment. This latter view is disputed among
economists, however, especially in recent contributions that question the validity
of the classical zero-tax results. One option to tax the normal return might be a
comprehensive business income tax, which taxes interest at the corporate level.
This would eliminate the inherent bias toward debt finance in current corporate
income tax systems, and it could eliminate the need to have complicated schemes
for taxing the normal return at the personal level. Given that such a system would
increase the tax base, it could come along with a lower tax rate to yield the same
corporate tax revenue.
Consensus does exist on the taxation of economic rents, which provides a
strong rationale for some type of corporate income tax. Moreover, there is gen-
erally broad support for the withholding function of the corporate income tax
for enforcing the income tax, most notably for developing countries where
administrative capacity is limited. These motivations do not, however, unequiv-
ocally support a corporate tax system as countries currently have it. More effi-
cient design would point, for instance, to some kind of rent tax—with the pos-
sibility to tax the normal return at the personal level. A corporate rent tax can be
based on a simple cash-flow tax system. Alternatively, it can be implemented by
allowing a deduction for the normal return on equity. To the extent that rent
taxes at the source are subject to spillovers, for example from profit shifting or
tax competition, they can also be based on a destination basis, which is generally
more robust to spillovers. These aspects are discussed in more detail in subse-
quent chapters.

REFERENCES
Atkinson, Anthony B., and Joseph E. Stiglitz. 1976. “The Design of Tax Structure: Direct versus
Indirect Taxation.” Journal of Public Economics 6(1–2): 55–75.
Banks, James, and Peter Diamond. 2010. “The Base for Direct Taxation.” In Dimensions of Tax
Design, edited by Stuart Adam, Tim Besley, Richard Blundell, Stephen Bond, Robert Chote,
Malcolm Gammie, Paul Johnson, Gareth Myles, and James Poterba, 548–674. Oxford, UK:
Oxford University Press.

©International Monetary Fund. Not for Redistribution

22 Corporate Income Taxes under Pressure

Boadway, Robin, and Neil Bruce. 1984. “A General Proposition on the Design of a Neutral
Business Tax.” Journal of Public Economics 24: 231–39.
Bradford, David F. 2004. “The X Tax in the World Economy.” NBER Working Paper 10676,
National Bureau of Economic Research, Cambridge, MA.
Brown, E. Cary. 1948. “Business Income Taxation and Investment Incentives.” In Income,
Employment and Public Policy, edited by Lloyd A. Metzler and others, 300–16. New York:
Norton.
Chamley, Christophe. 1986. Optimal Taxation of Capital Income in General Equilibrium with
Infinite Lives. Econometrica 54(3): 607–22.
De Mooij, Ruud. 2011. “The Tax Elasticity of Corporate Debt: A Synthesis of Size and
Variations.” IMF Working Paper 11/95, International Monetary Fund, Washington, DC.
De Mooij, Ruud, and Michael Keen. 2016. “Debt, Taxes, and Banks.” Journal of Money, Credit,
and Banking 48(1): 5–33.
Department of the Treasury. 1992. Integration of the Individual and Corporate Tax Systems:
Taxing Business Income Once. Washington, DC: Government Printing Office.
Feld, Lars P., Jost H. Heckemeyer, and Michael Overesch. 2013. “Capital Structure Choice and
Company Taxation: A Meta-Study.” Journal of Banking and Finance 37(8): 2850–66.
Hall, Robert E., and Alvin Rabushka. 2007. The Flat Tax. Stanford, CA: Hoover Institution
Press.
Hebous, Shafik and Alexander Klemm. 2020. “A Destination-Based Allowance for Corporate
Equity.” International Tax and Public Finance 27: 753–77.
International Monetary Fund (IMF). 2009. “Debt Bias and Other Distortions: Crisis-Related
Issues in Tax Policy.” IMF Policy Paper, Washington, DC.
International Monetary Fund (IMF). 2016. “Tax Policy, Leverage and Macroeconomic
Stability.” IMF Policy Paper, Washington, DC.
Judd, Kenneth L. 1985. “Redistributive Taxation in a Simple Perfect Foresight Model.” Journal
of Public Economics 28(1): 59–83.
Luca, Oana, and Alexander Tieman. 2016. “Financial Sector Debt Bias.” IMF Working Paper
13/217, International Monetary Fund, Washington, DC.
Meade, James. 1978. The Structure and Reform of Direct Taxation. London: George Allen and
Unwin.
Mirrlees, James, Stuart Adam, Tim Besley, Richard Blundell, Stephen Bond, Robert Chote,
Malcolm Gammie, Paul Johnson, Gareth Myles, and Jim M. Poterba. 2011. Tax by Design.
Oxford, UK: Oxford University Press.
Straub, Ludwig and Ivan Werning. 2020. “Positive Long-Run Capital Taxation: Chamley-Judd
Revisited.” American Economic Review 110(1): 86–119.

©International Monetary Fund. Not for Redistribution