JATA

Vol. 27, No. 1

Spring 2005
pp. 91–99

What Corporations Say They Do,

and What They Really Do:
Implications for Tax Policy
and Tax Research
Joel Slemrod

W
hen Kaye Newberry asked me to address this group, I jumped at the chance for
two reasons, other than an excuse to hang out next to Walt Disney World with
everyone: (1) to offer an apology, and (2) to ask you some questions.
A meeting of accountants to talk about taxation sounds like the setup for any of a
thousand or so jokes. A meeting of accountants whose idea of an interesting lunchtime
speaker is an economist should be the setup of a thousand more. Many of these jokes
concern what a boring lot we are. The somewhat enlightened believe that accountants and
economists aren’t boring people, but just get excited over boring things. Even within our
respective academic disciplines, tax academics don’t get a lot of respect. Of course, the
truly enlightened know that both stereotypes are untrue. I will not stoop to reply to the
claim that we are boring people—after all, here we are at Walt Disney World, aren’t we?
That taxation is a boring topic is demonstrably rubbish. In reality, it is boring only to those
who are uninterested in human nature and in the institutions that have been designed to
channel human nature to build productive civilizations. Taxation is about the relationship
between individuals and the state, about honesty and trust, and about how a society designs
institutions to try to overcome the free-rider impulse that threatens to undermine beneficial
collective action.
Two of these institutions, the public corporation and the corporation income tax, are
under attack these days. Enron, WorldCom, Tyco, and others have caused a rethinking of
the governance of public corporations, and a new set of laws. Corporate tax collections as
a percent of GDP or total federal taxes have been in a steady decline for four decades. In
2003 there was a substantial cut in the taxation of dividends, and there are rumors that a
second Bush administration could entertain a serious proposal to replace the income tax
with a consumption-based tax.

Joel Slemrod is a Professor and Director of the Office of Tax Policy at the University of
Michigan.
This article is based on remarks made at the ATA luncheon in Orlando, FL on August 9, 2004. I am grateful to
Sanjay Gupta, Michelle Hanlon, Lillian Mills, George Plesko, and Terry Shevlin for comments on an earlier draft;
they are not, though, in any way responsible for the views expressed herein.
Editor’s note: Professor Slemrod’s presentation was extremely well received and his opinions and insights warrant
wider dissemination among the readers of JATA. To this end, I invited Professor Slemrod to summarize his remarks
in this article.

91
92 Slemrod

How we tax corporations is the topic I have chosen to discuss today. Has the corporation
tax changed from being a tax, in the sense of an extra burden, into a shelter? If so, what
should be done about it? This second topic will lead naturally to addressing the case for
increased disclosure of tax information as well as greater conformity between tax and book
definitions of income, and to thinking carefully about the unique characteristics of public
corporations. I will argue that answering these questions requires that accountants and
economists work together.
Is the corporation tax a burden? It is likely that for all but the biggest companies, the
answer is now no. With the existence of S corporation status, limited liability companies,
and other hybrid entities, for businesses that are not and will never be public, the corporation
tax is an option for reducing tax. All of the nontax aspects of alternative business organi-
zational forms (except being publicly held) can be achieved while being subject to the
corporation income, or not being subject to it. So, compared to not existing, the corporation
income tax is a tax-reducing option, primarily because of the 15 percent and 25 percent
tax rates levied on taxable income under $75,000. Indeed, a natural life cycle of a business
is for it to be a pass-through entity in the early years of operation when losses are likely,
and then to convert to C corporation status when it achieves consistent profitability.
From a public policy standpoint, why allow businesses other than public corporations
to be taxed as C corporations? Why not require pass-through?1 Graduation of corporation
tax structure rates mimics the graduated rate structure of individual income tax, but cannot
be similarly defended as a way to impose a progressive distribution of tax burdens. Indeed,
I suspect that the average wealth of a small business owner is higher than the average
wealth of someone who owns shares in a giant public corporation, so that the graduation
of the corporation income tax structure may actually erode the progressivity of the overall
tax burden.
So the corporation tax system, as an extra burden, is now essentially for public cor-
porations only. It may, at the margin, be a disincentive to being public, but it is no longer
a disincentive to being a corporation. Note the connection of this argument to one possible
effect of the Sarbanes-Oxley bill, which, by making the reporting requirements and accom-
panying penalties more onerous, may cause some public companies to go private and deter
some private companies from ever going public. Thus, like the corporation income tax
itself, Sarbanes-Oxley may affect the margin of being a public company.
This raises an interesting public policy question: Why tax public corporations differ-
ently from other businesses? Is it an accident, or an evolution to a sound policy? One factor
is that it is infeasible to tax them as pass-through entities. Although this may be literally
true, we could get much closer to that model through a partial integration system that treats
the corporation tax essentially as a withholding tax and offers credits to shareholders for
taxes corporations pay on their behalf.
Our system of taxing public corporations is often derided as an especially onerous and
inefficient double tax. With that view as justification, the 2003 tax act sharply reduced the
taxation of corporate income by reducing the maximum personal tax rate on dividends to
15 percent.
But do we really tax corporations much at all? Although we still collect substantial
money from the corporation income tax, it is not nearly as important a revenue source as
it was a few decades ago. Revenues have recently been about 1.5 percent of GDP, compared
to 5 percent in the 1950s, 4 percent in the 1960s, 3 percent in the 1970s, and 2 percent in

1
Indeed, something close to this was suggested in The American Law Institute (1999).

The Journal of the American Taxation Association, Spring 2005

What Corporations Say They Do, and What They Really Do 93

the 1980s and 1990s. Why the decline? The standard answer, due to a previous work by
Auerbach and Poterba (1987), is that the biggest factor is a decline in true profits, not in
the effective tax rate imposed on those true profits. Surprisingly, this striking conclusion
has sparked little further research.
In addition, several studies have documented a large and—at least until re-
cently—growing gap between the book income reported on public corporations’ financial
statements and the taxable income of corporations, a gap which remains even after taking
account of known differences in the accounting procedures used for book and tax income.
(See, for example, Manzon and Plesko [2001] and Mills et al. [2002].) This phenomenon
has prompted some observers to conclude there is growing tax avoidance.
The nature of the tax base that produces what is left of corporation tax revenues is also
in doubt. In recent work with Gordon et al. (2004) I investigate the extent to which the
U.S. corporate income tax system collects tax on capital income, as opposed to pure profits
or labor income shifted into the corporate tax base. We estimate how much tax is collected
on capital income by calculating how much tax revenue would change if the tax system
were modified to exempt income from capital in present value—sometimes called an ‘‘R-
base tax’’—while leaving the tax rate structure and tax incentives otherwise unchanged.
The difference between actual revenue and revenue under this alternative tax system is a
measure of how much tax on capital income is collected under current law. We find that,
as of 2004, the U.S. tax system has returned to the situation of the mid-1980s wherein our
income tax system raises little revenue from taxing capital income. A corporate R-base tax
would collect only $30 billion, or about 15 percent less than the existing corporate income
tax, so very little of the corporate tax base is capital income, as assumed in most economic
models of the tax. We’re taxing something, but not capital income.
What to do about this state of affairs? Some would say drop the income tax completely,
arguing that income is inherently difficult to measure, and even if it could be measured, on
efficiency grounds an income tax is inferior to a consumption tax. The business tax com-
ponent of a consumption tax can be implemented either only on retailers, as in a retail
sales tax, or on all businesses as in a value-added tax (under which labor is not deductible),
or as part of a Hall-Rabushka type flat tax (under which labor is deductible). Under these
systems there is no pretense of measuring annual income. So conformity of book and tax
definitions of the tax base is out of the question, although there still may be information
gained by comparing conformed definitions of items such as sales.
Others say that, rather than abandon the income tax, the tax—and especially the cor-
porate income tax—should be strengthened. Initiatives along these lines follow several
tracks. Two of these tracks, greater disclosure and greater conformity between tax and book
income, I will discuss in a minute.
Before I do, I want to say a few words about the accounting and economics approaches
to tax research. I’ve already argued that the U.S. corporation tax has become a tax in the
sense of a burden only on public corporations, on that special institution that has played a
key role in allowing large business organizations to raise capital and in allowing investors
to expand and diversify wealth holdings. Of course, its public nature also means that ef-
fective control of the corporation is not exercised by the legal owner but instead by hired,
professional managers. It is inevitable that these managers know much about the business
that most shareholders do not, and potentially self-interested behavior by these managers
means that pressures from external investors, and perhaps formal and informal contracting
arrangements, are needed to induce managers to pursue shareholder-value-maximizing
behavior.

The Journal of the American Taxation Association, Spring 2005

94 Slemrod

Which brings me to accountants: objective and verifiable accounting information fa-

cilitates shareholder monitoring of business operations, and supplies a wide array of con-
tractible variables for determining the financial rewards to managers from incentive plans
designed to align the interests of shareholders and managers. You, of course, know that.
I am not an accountant, but among public finance economists, I travel in your circles
more than most economists. For the last 15 years I have been participating in an annual
meeting designed to bring together academics from accounting and economics to present
and discuss work-in-progress on taxation. Over these 15 years I have thought quite a bit
about the comparative advantages and interests of accountants and economists and have
formulated some gross generalizations about the differences. Here are two:

● Economists are more motivated by normative, or policy, questions.

● Economists are more motivated by the behavioral response to tax policy. We are
interested not only in how profitable a corporation is, but also in the actions it has
taken to make it profitable.

Indeed, the behavioral response to taxation is at the core of the modern economics of
taxation and underlies how we economists do our normative evaluation of tax systems in
terms of equity and efficiency. Behavioral response matters for assessing equity because
we need to know about it in order to assess the incidence of taxes—who really bears the
burden and whether the statutory burden is shifted due to price changes. This is especially
important and difficult for corporation tax, because it is meaningless to say corporations
bear the burden of taxes. Behavioral response matters for assessing efficiency because, in
most situations, the responses that taxes induce are symptoms of economic waste.
I have written in the past that it is useful to classify the behavioral responses to taxation
into three types: (1) real responses like saving, investment, and labor supply, (2) responses
that reshuffle, relabel, or recharacterize the real activities of the taxpayer, and (3) timing
responses that take advantage of changing tax rules or rates. (See, for example, Slemrod
[1992].) I have argued that these responses generally form a hierarchy with respect to the
magnitude, or elasticity, of behavioral response, with timing responses being the most elastic
and real responses being the least elastic.
I now have an apology to make. I used to think that there was another distinction
between accountants’ focus and economists’ focus: that accountants mostly study what com-
panies say they do (and how they are doing), while economists mostly study what compan-
ies actually do. In my hierarchy writings I’ve characterized this as the difference between
an accounting or renaming elasticity, and a real elasticity. Examples of renaming responses
include: in response to an investment tax credit, switching from immovable internal walls
to movable (and creditable) walls; in response to an R&D credit, suddenly realizing that
some of the things you were doing are research; or reconfiguring, in order to recharacterize
for tax purposes, a security as debt rather than equity.
My apology is that, at an unconscious level, this labeling may have been a prioritization
as well; as in we economists are ‘‘real’’ men (or women) or what we study are the ‘‘real,’’
as in important, issues—real things such as investment and R&D.
I now see the issue differently. I now see that one of your comparative advantages and
interests is the study of how the institutions related to public corporations address the
inherent and crucial information asymmetry problems. I should have known better. After
all, information asymmetry is at the core of normative public finance, known as optimal
tax theory. Each year I tell my graduate students that, if the IRS could costlessly observe

The Journal of the American Taxation Association, Spring 2005

What Corporations Say They Do, and What They Really Do 95

everything about taxpayers, taxes could be collected with no excess burden because indi-
vidual tax liability could be based on an inalterable correlate of individual well being, call
it ability. It is because ability cannot be measured well that governments are left to consider
taxing on the basis of alterable second-best correlates of well being such as income, con-
sumption, or wealth. Because they are alterable, taxing them causes behavioral response
and therefore inefficiency. Thus, measurement and information are at the core of the nor-
mative economics theory of taxation. And information is at the heart of the problem of
public corporations.
Publicly owned companies have to say how they are doing twice, to two different
audiences who have a stake in assessing how much income is being generated. One is a
public accounting, to shareholders, potential shareholders, creditors, and potential creditors.
Rules are set by the SEC directly and through SEC oversight of private standard-setting
bodies such as the FASB. The other is a nonpublic accounting, to the IRS. Like most
shareholders, the IRS takes a portion of income without participating in management, and
to monitor that it is receiving its appropriate share, it receives confidential, and possibly
misleading, reports on how the business is doing.
One class of information-related initiatives designed to respond to the recent corporate
scandals has to do with increased disclosure to each of the two audiences. One kind of
disclosure initiatives requires more disclosure to the IRS. Examples are disclosure require-
ments of certain transactions linked to abusive sheltering, and the new M-3 form that
requires more detailed information concerning the reconciliation of book and tax income.
The second class of initiatives being discussed is increased disclosure about corpora-
tions’ tax situation to the public. In this case, more disclosure means some disclosure
because, although financial statements contain some tax-related items, it is generally diffi-
cult, if not impossible, to infer a company’s true tax situation. I believe that public disclosure
of the entire corporate tax return is unwise social policy, because it would compromise
proprietary information and dilute the quality of information reported to the IRS. But I do
think that public release of tax liability and taxable income, as has been required in two
separate periods in U.S. tax history, would enhance the transparency of public policy.2 There
is a potential downside, though, as it might facilitate incentive contracts for tax managers
by making it easier to benchmark their performance against the performance of tax de-
partments in the same sector.
Public disclosure of some tax return information might also be informative for investors.
For one thing, it might help to understand the underlying profitability of the corporation.
But it might also exacerbate the race to the bottom of effective tax rates (ETRs). After all,
as a shareholder, although I do not want the company to be cheating me, I may want it to
cheat the IRS, at least cheat them optimally, taking into account the chances of being caught
and any attendant penalties. There is evidence that lower ETRs improve company valuation.
Where do shareholders think these lower ETRs come from? Having smarter people in the
tax department than other companies? To some extent that may be true, but part of it, savvy
investors realize, is that lower ETRs result from a more aggressive stance that pushes the
limits of what is legal. This is no different with regard to other regulations, including
environmental regulations, and so on. To be sure, there are investors out there who place
value on corporations playing strictly by the rules, but I am aware of no evidence that
investors are willing to accept a lower after-tax rate of return to investing in tax-compliant

2
The pros and cons of more public disclosure of corporate tax return information are discussed in Lenter et al.
(2003).

The Journal of the American Taxation Association, Spring 2005

96 Slemrod

firms, the way some investors apparently will accept a lower return on companies who are
enlightened with regard to the environment or the use of child labor.
A recent paper by Desai et al. (2003) argues that the connection between tax evasion
and managerial shenanigans goes deeper. They argue that the kinds of behavior that cor-
porations undertake for tax-planning purposes facilitate managerial diversion—okay, steal-
ing—from shareholders by making business operations more convoluted and less transpar-
ent. This implies that more effective enforcement of tax may have positive spillover effects
for corporate governance.
Let me say a few words about conformity. Why not harmonize the definition of income
for book and tax purposes? Then if a company tells the world it earned an extra dollar, it
owes tax on another dollar of income; if their crack tax department manages to reduce
income for tax purposes, they have to admit, and probably try to explain away, the lower
income to the world.
As a practical matter, complete conformity with the sole goal of accurate income mea-
surement is a nonstarter in this country. Tax policy makers will never relinquish the ability
to use the tax system to reward constituencies and incentivize behaviors. It is the same
reason that Milton Friedman predicted decades ago that we would never have a clean and
simple income tax base. In addition, recall the push toward a consumption tax base, which
would completely sever the connection between the business tax base and income.
Conformity is not, however, an all-or-nothing issue. Indeed, I find that the idea of, in
some way, tying tax liability to publicly stated income is an intriguing one. Indeed, this
was the law for a few years in the late 1980s when book income figured in the AMT base.
The evidence on whether this affected reporting of earnings is, as I read it, somewhat mixed.
There is no evidence that I know of concerning how it affected what companies did, as
opposed to what they say they did.
Recognizing the centrality of information to the analysis of public corporations touches
on one of the principal justifications for the corporation income tax—that it is an efficient
way to withhold tax on behalf of shareholders. Bird (1996) of the University of Toronto
stated this argument most succinctly when he wrote: ‘‘The key to effective taxation is
information, and the key to information in the modern economy is the corporation. The
corporation is thus the modern fiscal state’s equivalent of the customs barrier at the border.’’
Of course, who remits tax to the government and what triggers the liability are two
conceptually separate matters. After all, businesses remit most tax based on labor income
in the U.S., too. In either a completely integrated or a classical system of taxing corpora-
tions, the point of collection could be either the corporation or the shareholder. But it has
long been a no-brainer among tax administrators that it is more efficient to collect tax from
a small number of large corporations than from a large number of shareholders.
The apparent proliferation of corporate tax shelters challenges the notion that the cor-
poration is the more efficient point of collection. If there are economies of scale in the
consumption of tax shelters, then collecting tax revenue from big, public corporations might
not be that much more efficient than collecting it from shareholders, after all. Indeed,
because of the possibility of information reporting by corporations regarding dividends and
matching this information with shareholders’ income tax returns, dividend receipts are ar-
guably much easier to monitor than corporation income.
Resolving this issue brings us back to the separation of ownership and decision making.
Consider this question: would a corporation be as tax-aggressive if the tax remitted by the
corporation was unrelated to the company’s tax aggressiveness (related to book income?),
so that any tax savings from aggressiveness accrued directly to the shareholders by lowering

The Journal of the American Taxation Association, Spring 2005

What Corporations Say They Do, and What They Really Do 97

their personal tax liability? If the answer to this question is yes, they would be equally tax-
aggressive, then the point of collection is immaterial to tax shelter policy. If, though, the
answer is no, then tax savings that accrue directly to the shareholders would not, on average,
be pursued quite as aggressively by public corporations, and devolving remittance respon-
sibility to shareholders might inhibit aggressiveness. It also suggests that the move taken
in 2003 toward lowering taxes on dividends—which in a sense transfers the responsibility
for remittance of taxes on corporate income away from shareholders—is a move in exactly
the wrong direction.
Of course, there is information asymmetry between corporate insiders and shareholders
not only about the company’s profitability and prospects, but also about the aggressiveness
of the tax positions taken. The shareholders do not make decisions about accounting and
taxes directly, but through their agents, the chief financial officer or the vice president for
taxation. In order to align the incentives of the decision makers and the shareholders, the
managers are likely to have been granted stock options and may also have other ways their
compensation is tied to observable outcomes. This situation suggests some fascinating ques-
tions. How does the contract offered to the tax decision maker change when there are
penalties imposed by the IRS or SEC? Does the contract change to offset the intended
purpose of the penalties? From a normative point of view, what is the most appropriate
cocktail of tax evasion penalties to be imposed on the company and directly on the managers
within the corporation, knowing that the contract may very well adjust to the enforcement
regime?
Some of my own recent work with Keith Crocker (Crocker and Slemrod forthcoming)
explores this set of questions by examining corporate tax evasion in the context of the
contractual relationship between the shareholders of a firm and a tax manager who possesses
private information regarding the extent of legally permissible reductions in taxable income,
and who may also undertake illegal tax evasion. Using a costly state falsification framework,
we characterize formally the optimal incentive compensation contract for the tax manager
and, in particular, how the form of that contract changes in response to alternative enforce-
ment policies imposed by the taxing authority. The optimal contract may adjust to offset,
at least partially, the effect of sanctions against illegal evasion. We find a new and policy-
relevant nonequivalence result: penalties imposed on the tax manager are more effective in
reducing evasion than are those imposed on shareholders. The intuition is straightforward:
the penalties imposed on the company affect behavior only to the extent they are passed
on to decision makers via the incentives in the compensation contact. This dilutes the
deterrent effect of the penalties.
I’ve been talking mostly about theory and public policy. Let me close with a few
thoughts about empirical analysis. Empirical analysis of how corporations respond—in what
they do, as opposed to how they report on what they do—to the tax system that is based
on financial reports is hampered by a double whammy. First, the financial reports may be
massaged to mislead the shareholders. Second, and more prosaic, is the fact that corpora-
tions do not in their financial reports have to reveal exactly what tax they pay to the IRS,
much less what their ‘‘true’’ taxable income is. Some of us are lucky enough to get a peek
at sanitized samples of corporate tax returns. In this case, we are not so naı̈ve as to take
these reports as the unalloyed truth. Both the tax return and the financial statement are
snapshots of what the taxpayer wants some audience—either the IRS or the investing pub-
lic—to see. We may be misled if we ignore this and assume the reports to be the unvar-
nished truth. This is obvious to those of you who study financial accounting. But it is less
obvious to those of us who study the affect of the tax system on what corporations do.

The Journal of the American Taxation Association, Spring 2005

98 Slemrod

These issues have important implications for a perennial focus of economists: the effect
of the tax system on corporate investment. We first calculate the tax-adjusted cost of capital
and then examine how changes in that cost of capital affect investment. Investment numbers
for the National Income and Product Accounts, used in aggregate time-series studies of the
effect of tax on investment, come from Census Bureau surveys. Maybe companies tell the
truth when the data reported is confidential and has no consequences either for tax liability
or for anyone’s perception of the profitability and/or market value of the company. But
micro-level studies of the effect of taxation on investment have used the capital expenditure
variable from companies’ uses of funds statement. I’d like to know if these figures are
subject to systematic bias related to tax avoidance or accounting malfeasance.
More generally, I’d like to know more about the relationship among tax avoidance,
accounting malfeasance, and the incentive to invest. The modern theory of the taxable
income elasticity argues that the sum of the behavioral responses is all that matters for
assessing the efficiency of taxes—the higher the total behavioral response elasticity, the
more inefficient a tax is—and the breakdown is not critical (see Feldstein 1999; Slemrod
1998). But that is not the end of the story. For one thing, the behavioral responses interact.
My favorite example of this interaction arose during the recent debate over the estate tax,
during which opponents of estate tax often made two points—that it inhibits wealth accu-
mulation, and it is easily avoided. At first blush, these statements seem to be mutually
exclusive—why bother to reduce wealth accumulation if the estate tax can be avoided, or
evaded so easily? The same question applies to investment. Is the gain from corporate tax
avoidance inframarginal, or does it reduce the marginal effective tax rate on new invest-
ment? The answer depends on whether investment is what I call avoidance-facilitating. If
more investment (i.e., scale) allows the corporation to effectively ‘‘get away’’ with more
sheltering, then there is an implicit subsidy to investment.
Another reason to distinguish among the behavioral responses is that ‘‘real’’ responses
depend on largely immutable aspects of people’s preferences and firms’ technologies, but
avoidance or accounting responses can often be intermediated by a host of policies. How
much labor supply changes in response to the effective real wage rate depends mostly on
individual preferences, but to what extent companies invert and reincorporate in Bermuda
depends on the rules and regulations that govern such behavior. In the latter case, there is
an appropriate enforcement response that in effect determines an ‘‘optimal’’ elasticity of
the behavior that trades off the costs of limiting behavioral response against the protection
of the revenue base. The response elasticities are not immutable parameters that must be
accepted in the formulation of tax policy, but are themselves parameters than can be altered
by policy.
I said at the outset I had an apology to make and some questions to ask. I’ve made
my apology, and now it’s time to ask my questions. My apology was an acknowledgement
that economists who study tax need to pay more attention to the issues that accountants
have been studying for decades. There are important and profound interactions between
what corporations say they do (and how they are doing) and what they actually do. We
need to think about developing theoretical models that address this reality and empirically
test how these interactions play out. My questions are designed to stimulate your thinking
about the kinds of questions that economists have been pondering for decades:
1) How is the incentive to make new investments (or other real behavior) affected by
tax sheltering? Are do-it-yourself tax cuts as effective as legislated ones?

The Journal of the American Taxation Association, Spring 2005

What Corporations Say They Do, and What They Really Do 99

2) How is the incentive to invest affected by the amount of accounting earnings, as

opposed to actual earnings, it produces? Do investments that provide more book
earnings effectively have a lower hurdle rate?
3) Does accounting manipulation or tax avoidance affect statements of perennial ob-
jects of study by economists, such as investment?
4) Can we integrate models of governance and tax, both of which are inherently issues
of information asymmetry?
We live in an exciting time when the corporation and how we tax it are being re-
evaluated. It is our job to inform the public policy debate. (Remember I said that economists
are concerned about public policy.) This is a challenge because the data we have is not
necessarily designed to inform, but is often designed to mislead. To make progress, we
need your expertise in information so that our policy advice is more informative than
misleading.

REFERENCES
The American Law Institute. 1999. Taxation of Private Business Enterprises. Reporters’ Study by
G. K. Yin and D. J. Shakow. July. Philadelphia, PA: The American Law Institute.
Auerbach, A. J., and J. M. Poterba. 1987. Why have corporate tax revenues declined? In Tax Policy
and the Economy, Volume 1, edited by L. H. Summers, 1–28. Cambridge, MA: NBER and MIT
Press.
Bird, R. M. 1996. Why tax corporations? Working Paper 96-2, December. Prepared for the Technical
Committee on Business Taxation, Canada.
Crocker, K., and J. Slemrod. Year. Corporate tax evasion with agency costs. Journal of Public Eco-
nomics (forthcoming).
Desai, M. A., A. Dyck, and L. Zingales. 2003. Corporate governance and taxation. Working paper,
University of Chicago and Harvard Business School.
Feldstein, M. 1999. Tax avoidance and the deadweight loss of the income tax. Review of Economics
and Statistics 81: 674–681.
Gordon, R., L. Kalambokidis, J. Rohaly, and J. Slemrod. 2004. Toward a consumption tax, and
beyond. American Economic Review Papers and Proceedings 94 (2): 161–165.
Lenter, D., D. Shackelford, and J. Slemrod. 2003. Public disclosure of corporate tax return informa-
tion: Accounting, economics, and legal issues. National Tax Journal 56 (4): 803–830.
Manzon, G., Jr., and G. Plesko. 2001. The relation between financial and tax reporting measures of
income. Tax Law Review 55 (2): 175–214.
Mills, L., K. Newberry, and W. B. Trautman. 2002. Trends in Book-Tax Income and Balance Sheet
Differences. IRS Research Bulletin, Publication 1500. Washington, DC: Government Printing
Office.
Slemrod, J. 1992. Do taxes matter? Lessons from the 1980s. American Economic Review Papers and
Proceedings 82 (2): 250–256.
———. 1999. Methodological issues in measuring and interpreting taxable income elasticities. Na-
tional Tax Journal 51 (4): 773–788.

The Journal of the American Taxation Association, Spring 2005