Transcript R i

Unit 8 Seminar
Tax Incidence: Applications
Cost Benefit Analysis
The Harberger General Equilibrium
Model of Tax Incidence
Variation of typical two-goods (X, Y) and two-factors (K,L) general
equilibrium model
Individuals have identical utility functions over X and Y, goods
and factor markets are perfectly competitive, and there are no
pure economic profits for firms
It must also be assumed that Y is relatively capital-intensive and
X is relatively labor-intensive (required to guarantee the prices of
the goods and the factors vary as the economy moves from one
equilibrium to another along the frontier)
Perfect Competition
The assumption of perfect competition has two important
The capital and labor supplied to the market must always be
fully employed in equilibrium
• If there is unemployment, then production would fall until the
X and Y firms are willing to hire all of the labor
Individuals have to receive the same wages and same return to
capital whether they are used in the production of X or Y
• If not, then workers would move between X and Y firms,
driving up the wages in the firm they left and driving down
wages in the firm they entered, until equality was reached
Tax Incidence Equivalence
Because Harberger assumes that all tax revenues collected are
returned lump-sum, it holds that the incidence and the impact of
lump-sum taxes are identical
Examples include taxes on individual income earned from
labor supply or supply of capital and broad-based tax on
Haig-Simons income
Because the side of the market taxed does not matter:
1. A payroll tax on all firms’ use of labor is equivalent to an
income tax on wages received
2. A tax on all firms’ returns to capital is equivalent to an
income tax on the earnings from capital
3. A value-added tax on payments to labor and capital by all
firms is equivalent to a broad-based income tax
Tax Incidence Equivalence continued
Note: These factor taxes are all lump-sum so if revenue is returned
lump-sum, they do not affect prices or factor allocations between X
and Y firms (i.e. their incidence and impact is equal)
Equivalence between broad-based taxes implies equivalence
between an expenditure tax on both goods and a broad-based
income tax
A general sales tax levied on both the Y and X firms is equivalent to
a value-added tax on the payments to labor and capital by both
• None of these taxes has any effect on the economy if the
revenues are all returned lump sum to individuals
This means that the only interesting taxes, at least in terms of
analyzing tax incidence, are specific taxes that are distorting and
not lump sum
The Incidence of Corporation Income Tax
Harberger assumed that the relatively capital-intensive Y firms
were corporations and the relatively labor-intensive X firms were
non-corporate businesses
The corporate income tax makes K more expensive and raises
the marginal cost of producing Y
If Y and X are substitutes in consumption, this increases the
demand for X
Post-tax prices are assumed to be equal because the tax drives
the economy below its production possibility frontier so that the
marginal rate of technical substitution are no longer equal for X
and Y firms
The Incidence of Corporation Income Tax cont’d
- Supply of capital is assumed to be fixed
The corporate income tax causes the demand for capital by the Y
firms to shift down by the full amount of the tax
This reduction causes owners of capital to shift from Y to X firms
which causes capital supply curves to shift until the net rates of
return are equal
• Result is that all suppliers of capital lose (and government’s
desire to impose burden only on corporation stockholders fails)
The Incidence of
Corporation Income Tax cont’d
In this model, the incidence of the tax depends on:
The size of the tax
The importance of the taxed factor to the industry being taxed
• Both determine the degree to which taxed firms react to the
tax (i.e. how much the supply curves above move)
The elasticities of demand for X and Y
• Determines the degree to which shifting of resources occurs
between X and Y (i.e. how much quantity demanded of Y
decreases and demand for X increases)
The elasticity of substitution between K and L in the two
• Determines the slope of the demand curves for capital
The Incidence of
Corporation Income Tax cont’d
Harberger noted that the best empirical estimates of
the final three factors, at the time, suggested that
the return to capital would fall so far that capital
would suffer almost the entire burden of the tax
• Most economists agree that the burden of the
corporate income tax falls almost entirely on the
suppliers of capital
Variations in the Baseline Model
Variable factor supplies
Assuming factors of production are variable rather than fixed,
results in major taxes are no longer lump-sum
• This means that the incidence is no longer the impact
The biggest difference occurs in the long-run, in an overlapping
generations (OLG) model
• This is because the corporate income tax is borne primarily by
the suppliers of capital, the savings rate decreases in the
long-run which decreases investment, the marginal product
of labor and the real wage (which means that some portion of
tax becomes borne by labor)
Variations in the Baseline Model continued
Non-identical individuals
Affects individuals in a static context as a tax on good X that raises
the relative price ratio, forcing more of the burden onto those who
consume a higher proportion of X
In an OLG model, a switch from an income to an expenditure tax
places a disproportionate burden on the older/retired generation by
taxing them for a third time on their working years income. The three
taxes are:
1. They saved out of post-tax dollars
2. They paid taxes on the interest earned on savings
3. They now pay expenditure tax as they spend their savings
Variations in the Baseline Model continued
- Overall then, the supply tax has no effect on the
market for capital
– It does impact local owners of capital though, as the netof-tax-price decreases by the full amount of the tax so
that they suffer the full burden of the tax
- A capital income tax on firms within the locality (a
demand-side tax) shifts the demand curve for capital
down by the full amount of the tax
– Here the equilibrium amount of capital decreases, the
price of capital to firms increases, and the return top
owners of capital remains the same
Variations in the Baseline Model continued
- This demonstrates that the tax-on-demand side
differs from the tax-on-supply side in two important
(1) The demand-side tax does not burden local owners of
capital like the supply-side tax does
(2) Only the demand-side tax lowers the amount of labor
and capital employed in the state or locality and the
amount of output produced (i.e. is “anti-business”)
Non-Competitive Markets
Harberger argues that under perfect competition the burden of the
corporate income tax is borne largely by stockholders
- Large, price-setting oligopolies are found almost exclusively in the
corporate business sector
Harberger’s assumption of perfect competition is probably not
accurate for effect of corporate income tax in such cases
- Economists have not settled on a general theory of oligopoly
- Based on existing theories, it is possible that managers of large
corporations can avoid some of the burden of a corporate income tax
by exerting their market power
• If the corporation is initially maximizing market share by
producing above profit-maximizing output, stockholders can
avoid the burden of the tax by insisting the firm cut back on
production and raise prices
The Sources and Uses Analysis of Incidence
Developed by Joseph Pechman of the Brookings Institution
Designed to exploit the detailed data sets on individuals and
families that various government agencies were collecting and
making available to researchers
– Approach used a variety of assumptions and results in the
literature to allocate the burden of the major U.S. taxes to the
different sources and uses of income of individuals and
families in the databases, before looking at income deciles to
determine whether taxes were regressive, proportional, or
Sources defined as transfers received and also earned income,
consisting of income from labor, capital, and land
Uses defined as consumption and saving
The Sources and Uses Analysis of Incidence
The allocation of tax burden to sources and uses
matters because:
1. Transfers disproportionately go to the poor
2. Income from capital disproportionately goes to the
3. The ratio of consumption to income falls sharply as
income rises
Tax burdens allocated to transfers and consumption
tend to be highly regressive while tax burdens
allocated to income from capital tend to be
The Major U.S. Taxes
Federal and state personal taxes
• Assume that the supplies of labor, capital, and land
are fixed so that the impact of personal income taxes
are their incidence
– Find that the federal personal income tax is highly
progressive over lower ranges of income due to
personal exemptions and standard deductions
– Tax is still progressive at higher incomes but the degree
is reduced due to exclusions and deductions available
to taxpayers who itemize
The Major U.S. Taxes continued
Social security payroll tax
Corporate income tax
State general sales tax
Local property tax
Social Security Payroll Tax
Levied on both sides of market with employers and
employees both paying a tax of 6.2% up to a
maximum of $90,000 in 2005 for individual incomes
earmarked for social security
Employers and employees pay an additional tax of
1.45% of income, with no upper limit, which is
earmarked for Medicare
Because the supply of labor is fixed the incidence of
the tax is the impact
Social Security Payroll Tax continued
The market does not allow the tax to be split,
however, because the supply of labor is perfectly
inelastic and the full burden of the tax falls entirely on
This makes the tax highly regressive because:
1. The taxes are allocated only to wages and income (i.e.
the capital portion of income escapes the tax)
2. Due to the $90,000 upper limit T/Y falls steadily as
income increases
Corporate Income Tax
Adopt Harberger conclusion that corporate
income tax is borne entirely by corporate
State General Sales Tax
- Typically levied at a single rate on a broad range of
manufactured goods
- Pechman/Okner assume that the supply of goods is
sufficiently elastic in the long-run to so that it can
reasonably be viewed as perfectly elastic
– This means that although tax is levied on firms, they can
pass all of tax onto consumers in the form of higher
prices (which eventually rise by full amount of tax)
- This makes the tax highly regressive
Local Property Tax
Pechman/Okner assume that the supply of land is fixed so that
landlords bear the entire incidence of the tax, equal to the land
portion of the property tax payment
– Because land is disproportionately owned by the rich, this
makes the property tax highly progressive
Overall, Pechman/Okner view the U.S. tax system as a combination
of progressive and regressive taxes that in their totality end up not
redistributing very much wealth
– According to their estimates, tax burdens vary from 20.6% at
the lowest decile of income to 27.1% at the highest decile of
Caveats to the Sources and Uses Approach
John Whalley (1984) cautioned that sources and
uses approach is very sensitive to assumptions
that are made
• By carefully changing assumptions, each of which
seem plausible, it is possible to make any tax
structure appear either progressive or regressive
The Long-Run Perspective
Switching from Pechman/Okner’s annual perspective to a lifetime
perspective changes the analysis over two important dimensions
1. The sources and uses of income differ across time periods
2. The two most important sources of lifetime income (transfers
and labor earnings) vary much less than their annual
Nonetheless, most lifetime studies reach the same conclusion as
Pechman/Okner’s annual studies: the U.S. tax system is not very
Sources and Uses Versus
General Equilibrium Modeling
In the 40 years since Harberger’s simple two-good two-factor
economy, significant advances have been made in the ability to
conduce general equilibrium modeling
– As with the sources and uses approach, however, these
advanced models rely importantly on the assumptions they
Accordingly, there is no definite, best model for analyzing tax
– The general equilibrium approach seems to be preferred by the
majority of public economists who study tax incidence, but the
sources and uses approach seems to dominate the public
Durability and
the Present Value Formula
• Private and government investments represent
additions to the stock of durable capital
– Capital constructed or purchased today is expected to
be productive and yield a stream of income for more
than one year
• The problem in accounting for such investments is that
dollars received in different years are not equivalent …
Compounding to Future Value
Suppose you spend $1 today to buy a U.S. Treasury note that
entails no risk and pays a rate of return of 8%
- Assuming no inflation, one year from today you will be paid
$1(1+.08) = $1.08
- You are indifferent between receiving $1 today and $1.08 one
year from today
The same holds for all future time periods:
$ value 1 year from today
= $1.08 today
$ value 2 years from today
= $1.17 today
$ value 3 years from today
= $1.26 today
$ value n years from today
This calculation is referred to as compounding to future values
because the result indicates what $1 today would be worth in
future time periods
Discounting to Present Value
- One can also calculate in the opposite direction (i.e.
how much would $1 at some time in the future be
worth today)
$ value 1 year from today
= $0.93 today
$ value 2 years from today
= $0.86 today
$ value 3 years from today
= $0.79 today
$ value n years from today
• This calculation is referred to as discounting to
present value because the result indicates what $1 in
future time periods would be worth today
The Present Value Formula
- Suppose a firm is considering whether to buy a
machine. Let:
= the initial investment cost
= the net revenues observed in year I
= the discount rate
= the last year the machine is projected to
generate a positive revenue
- The present value of the investment is:
PVi = -I0 + ΣNi=1(Ri/(1+ r)i)
The Present Value Formula continued
The firm’s investment rule is:
Invest if
PV > 0
Do not invest if
PV < 0
If a firm does not have sufficient funds to invest in all projects with
positive present value then it should invest in those projects that
provide the maximum present value
1. The discount rate is the opportunity cost of capital
2. The internal yield is the discount rate that sets PV = 0
3. Inflation does not affect the present value formula
4. Initial investment costs often occur over multiple years and must
also be discounted
Cost-Benefit Analysis vs. Private Investment Analysis
The biggest difficulty will be in determining the
different components of the present value
formula: the appropriate discount rate, r, and the
appropriate projected future stream of net
returns, Ri
Private Investment Analysis
- For a private firm, r depends on the saving
opportunity available to the firm
– The problem is that the true value of r is likely not known
to the firm
• The issue then is how the firm sets its guess as to the
true r
– If the guess is too low, the firm over-invests
– If the guess is too high, the firm under-invests
Private Investment Analysis continued
- For a private firm, Ri depends on the firm’s estimates
of the future, which cannot be known with certainty
– If the firm expects the economy to perform well over the
life of the investment, the Ri is higher and the firm is more
likely to invest
– If the firm expects the economy to perform poorly over the
life of the investment, the Ri is lower and the firm is less
likely to invest
Cost-Benefit Analysis:
Three Fundamental Principles
The principles for public managers are different but many of the issues
are similar
Three Fundamental Principles of Cost-Benefit Analysis
It is unrealistic to expect public policy makers to make cost-benefit
analyses that are exactly correct
• At best, cost-benefit analysis can only offer practical guidance
to policy makers
Policy makers should try to quantify the elements of the present
value formula to the extent possible for all potential government
• Policy makers should identify all sources of true project
benefits and costs, use state-of-the-art evaluation techniques,
and avoid attaching bogus benefits and costs
Cost-Benefit Analysis:
Three Fundamental Principles continued
(3) The cost-benefit analysis should assume that the
economy is at full employment, unless stated
specifically otherwise in the study
The main purpose of the study is to help policy
makers determine how to best make use of society’s
scarce resources
If anything other than full-employment is assumed,
then policy makers might overestimate the true
benefit of the project to society
The Elements of a Cost-Benefit Analysis
The public rate of discount
For private firms, the discount rate r clearly
represents the opportunity cost of capital,
because it is the rate of return that stockholders
could have earned on their savings
It is not so clear how to interpret the term for
public investment
The Elements of a Cost-Benefit Analysis
2. The opportunity cost view
Recall that (assuming X and M are roughly equal)
GDP = C + I + G
Under the assumption of full-employment, long-run GDP is assumed
to be constant
Because GDP is constant, the only way to have an increase in
public spending G is to have an offsetting decrease in
consumption C and/or private investment
The foregone return from the decrease in private investment is the
productivity of private investment on the margin
The foregone return from the decrease in private consumption is
the rate of return that individuals receive on their savings
The Elements of a Cost-Benefit Analysis
It follows from the previous slide that a government investment
must at least match this rate of returns to be worthwhile
This means that the appropriate discount rate for cost-benefit
analyses of public investment is a weighted average of the above
rates of return
– In practice, the appropriate rates of return and weights are
difficult to determine
– This is potentially problematic, because as noted above,
choosing values that are too high results in fewer investments
and vice versa
The Elements of a Cost-Benefit Analysis
3. Other views
- Public investments give rise to services that are
consumed by the public
– The appropriate discount rate depends on the way that
society views the marginal rate of substitution between
current and future consumption of such services (the
marginal social rate of time preference)
- The two different methods give rise to very different
values for the public discount rate and it is not clear
which is more correct …
The Elements of a Cost-Benefit Analysis
(1) Most economists agree that future consumption
should be discounted by society at a lower rate than
private savings
Savings entail the positive externality that future
earnings from savings provide future tax revenues
Current generation do not give enough weight to
future generations
Assuming a lower rate, address both of these
concerns by encouraging more public investment
The Elements of a Cost-Benefit Analysis
A popular theoretical model accounting for tax distortions
suggests that the appropriate formula is actually much more
complex than either of the above methods
Weitzman (2001) surveyed 2,160 economists in 48 countries and
found the mean recommended public rate of discount to be under
4% with a range of -3% to 27%
The U.S. Office of Management and Budget uses a rate of 7%
which appears closer to opportunity cost view
Problems in Measuring the Net Benefits
With regard to public spending, it is not possible to
use market prices as indicators of costs and
1. Intangibles
• Some aspects of the benefit of public project are
intangible and do not have a clear market price
• Examples: National security from national
defense, national prestige from moon landing, loss
of life to bridge construction workers, etc.
Problems in Measuring the Net Benefits
The common solution in such cases it to estimate
net benefits of the quantifiable elements of differing
public projects and assume that intangible benefits
must exceed the difference between those values in
order to make the project with greater intangibles
There is a great deal of debate as to how to quantify
the intangible cost of lives lost in construction of
public projects or lives saved by reduced pollution,
increased safety, etc.
Problems in Measuring the Net Benefits
2. Lumpy investments
Large government spending projects (hydroelectric
dams, light rail transit systems, etc) drastically
increase supply and decrease price
Economists prefer use Hicks equivalent variation
to measure such benefits because it provides a more
consistent ranking of alternatives by evaluating
change at original prices rather than new lower,
Problems in Measuring the Net Benefits
3. Non-marketed benefits and costs
- Many of the benefits for public spending projects are given away
rather than sold
– Lack of a market means there is no market price to attach to the
benefit to determine its value
4. Evaluating the source of the benefits
- Consider the benefit of building a highway in a rural area previously
served by rural roads
- The main benefit is that the interstate is safer and reduces commute
- The main question is how to appropriately evaluate the benefits of
increased safety and time savings
– One could use health care costs as an indicator of the benefit of
reduced non-fatal crashes and the average hourly wage as an
indicator of benefit of time savings, but it is not clear that these
are only aspects of such benefits
Problems in Measuring the Net Benefits
5. Hedonic price estimation
Consider the benefit of public spending projects that provide
cleaner air
• There is not a clearly defined market for clean air
The real estate market does incorporate the benefit of clear air,
in that when individuals purchase housing they are purchasing
all aspects provided by the house, including the house and
property itself, the characteristics of the surrounding
neighborhood and community, the amount of air pollution, etc.
Economists estimate the independent effect of each of these
aspects and use the estimate for air pollution
Problems in Measuring the Net Benefits cont’d
6. Contingent valuation
- Conduct a survey that asks people to attach a value to different
non-market benefits (i.e. ask how much people would be willing to
pay to preserve natural areas that they will never visit)
– Economists are mixed as to the value provided by such
7. Shadow prices
- Because various taxes create market distortions that cause
consumers and producers to face different prices, appropriate
weights (or shadow prices) must be assigned to the various prices
faced by consumers and producers
- Unfortunately, almost all markets are affected by distortions and
shadow prices are difficult to determine in practice
The Distribution of Benefits and Costs
- Incorporating end-results equity in a cost-benefit
analysis creates many difficulties as equity issues
are likely to overwhelm efficiency issues that lie at
the heart of cost-benefit analysis
- Nonetheless, distribution issues related to the costs
and benefits of public spending projects should be
Avoiding Some Pitfalls:
Bogus Costs and Benefits
- Politicians often have a vested interest in seeing
certain projects financed or not financed
– They may calculate bogus benefit-cost analyses to
try to slant the political process in their favor.
Techniques include:
1. Regional multipliers
2. The labor game
3. Double counting
Avoiding Some Pitfalls:
Bogus Costs and Benefits continued
1. Regional multipliers
- Large public projects require numerous support services while the
project is being built (restaurants, supermarkets, clothing stores,
homes, hotels, etc.)
– Such spending, or secondary benefits, are often included as
benefits in cost-benefit analyses
These secondary benefits are not a net benefit to society, as they
just represent a redistribution of resources across regions (i.e. there
are secondary losses to the regions that lose the resources)
– Because it is probably impossible to quantify how the secondary
benefits in one region compare to the secondary loss in another,
it is best to ignore such factors
Avoiding Some Pitfalls:
Bogus Costs and Benefits continued
2. The labor game
• Politicians often cite total wages paid to construction
workers as a benefit of the project
– This is ridiculous because the wages paid to workers
are part of the cost of the project not part of the
Avoiding Some Pitfalls:
Bogus Costs and Benefits continued
3. Double counting
Example: Suppose government is considering building a
new airport and wants to assess the cost that
additional noise pollution will have on citizens living
- Government might try to include both a contingent
valuation of how much people would pay to avoid
noise as well as the cost to homeowners of lower
housing prices
• Both of these measure the same thing meaning
that only one should be included in the analysis
Most economists agree that public policy debates
on government spending projects are well-informed
by careful cost-benefit analyses that:
1. Focus on true project costs and benefits
2. Attempt to quantify the various elements of the PV
3. Apply a consistent methodology to all potential projects
4. Avoid the inclusion of secondary benefits, employment
effects, and other bogus costs and benefits