Transcript Thomas A. Barthold, "How Economics Can Inform Tax Policy

FIVE KEY THINGS
How Economics Can Inform
Tax Policy Design
Thomas A. Barthold
Joint Committee on Taxation
National Tax Association
&
Office of Tax Policy Research
University of Michigan
Washington, D.C.
September 29, 2006
My Five Keys
• Economic incidence, not statutory
incidence rules.
• Short-run incidence may be different from
long-run incidence.
• The economic burden of a tax generally is
different from the tax’s revenue yield.
• Time value matters.
• Some costs are sunk and some costs are
variable.
1. Economic incidence, not
statutory incidence, rules
• Demand and supply determine the gross price of
exchange for goods, labor, and capital.
• A tax creates a wedge between the gross price
and the net receipt on an exchange of goods,
labor, and capital.
• How buyers and sellers respond to changes in
the gross price or net receipt determines a new
market outcome and who bears the tax.
-Implications
• Depending upon the elasticity of supply and
demand, it can be a fool’s errand to try to place
a burden on a specific party in the market.
 Is the burden of the payroll tax half on the employee
and half on the employer?
• Creating the appearance of imposing “the
burden” on the statutorily designated party can
create inefficiency from increased administrative
and compliance burdens.
 Could the administration of and compliance costs of
the credits for hybrid vehicles have been significantly
eased if it were a manufacturer’s credit?
2. Short-Run Incidence and LongRun Incidence May Differ
• In wake of sharp increases in gasoline prices after
Hurricane Katrina, several policy makers called for a
temporary repeal of federal motor fuels excise taxes.
• If the hurricane’s damage resulted in a reduced and
virtually immutable short-run supply of oil, the proposed
policy would have increased the profits of suppliers with
no change in price at the pump.
• However, over longer periods, economists generally
estimate that motor fuel taxes are borne largely by
consumers.
• Incidence is all about “elasticity,” behavioral response.
-Implications
• Policies may have different outcomes in
the long run than in the short run.
• The adage “an old tax is a good tax” need
not be valid as market conditions change.
3. Economic burden generally does
not equal revenue yield
• In addition to using resources to pay a tax,
market participants lose value from making
choices they would not have made in the
absence of the tax. The new choice is always
inferior to the pre-tax choice.
• Economic burden is the sum of resources
transferred to the government and the value of
distorted behavior.
• The greater the behavioral response to a price
change, the greater the economic burden,
regardless of the revenue yield.
-Implications
• Revenue is the wrong measure to look at when
assessing the burden and the distribution of the
burden of a tax.
 A low average tax rate does not necessarily mean a
tax does little economic harm.
 A high average tax rate does not necessarily mean a
tax does great economic harm.
• Because burden results from loss of resources
transferred to the government and changed
behavior, revenue gained with little behavioral
change is better than revenue gained with
greater behavioral change.
4. Time value matters
• A dollar today is more valuable than a
dollar tomorrow.
• True in real terms, more important in the
presence of inflation.
• Well-known concept, many times not
honored in tax policy design.
-Implications(1)
• Creates difficulty in defining income for the
tax base, particularly if the intent is to tax
real income.
E.g., Appropriate cost recovery schedules.
E.g., Realized capital gains.
E.g., Application of NOL, GBC, FTC
carryforwards.
-Implications (2)
• Seemingly different policies can create
identical economic incentives.
E.g., traditional vs. Roth IRA.
E.g., ITC vs accelerated depreciation.
• Subsidies delivered through Code cannot
be of equal value to all taxpayers unless:
Refundable, or
Deferred with interest.
5. Some costs are sunk and some
costs are variable
• A sunk cost is inframarginal, it does not
affect behavior going forward.
• A tax change may affect the value of an
existing asset (sunk cost), but generally
not its supply.
• Where an expense is variable, a tax
change may affect supply.
-Implications
• Policy makers spend significant time
considering transition relief.
For sunk investments such relief has no effect
on the economy.
Such relief is generally about perceptions of
fairness.
• Providing transition relief for sunk
investments can mean higher tax rates
and more distortion of variable
investments.