The Impact of Taxes on Saving, Investment,
and Economic Growth
Meeting of the President’s Advisory Panel
on Federal Tax Reform
San Francisco
March 31, 2005
Dr. Michael J. Boskin
T. M. Friedman Professor of Economics
Hoover Institution Senior Fellow
Stanford University
1
The Impact of Taxes on Saving, Investment, and Economic Growth

Real per capita income grows with productivity, which in turn rises with the
growth rate of capital per worker and the growth rate of technology;

Taxes affect the capital stock by affecting saving and investment;

Taxes may also affect hours of work and human capital investment;

Taxes, and some government spending, may affect the rate of advance in
technology;

Tax effects on human investment and/or technology can permanently affect
the growth rate of the economy;

Taxes on saving and investment affect the level of future income and hence
the interim growth rate; whether they permanently alter the growth rate
depends upon whether investment affects the rate of technical change, on
which economists disagree;

At the very least, we should strive for neutrality in the tax code toward
saving and investment.
2
National & Personal Saving in the Postwar U.S.
(% of GDP)
25.0%
20.0%
% of GDP
15.0%
10.0%
5.0%
Personal saving
National (gross) saving
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
1968
1966
1964
1962
1960
1958
1956
1954
1952
1950
0.0%
National (net) saving
3
National Income and Product Accounts (NIPA)
Measures of U.S. Saving
(% of GDP, alternative periods)
Gross national saving = personal saving + business saving + government saving
Net national saving
= gross saving - depreciation
Gross
national
saving
1951-60
1961-70
1971-80
1981-90
1991-2000
2000-04
Note: 1951-2004 average
20.9%
20.8%
19.7%
18.0%
16.5%
14.4%
18.8%
Net
national
saving
10.5%
10.9%
8.6%
5.9%
4.7%
2.1%
7.7%
Personal
saving
5.5%
5.9%
6.9%
6.4%
3.5%
1.2%
5.3%
4
Issues In The Measurement And Interpretation
Of The Low And Declining Saving Rate

The exclusion of capital gains (losses) means NIPA saving is not
synonymous with wealth accumulation
 Wealth effect of rising equity and home prices accounts for perhaps
half the decline in personal saving;
 Stock market gains decrease firm contributions to DB pension plans;
higher DB plan benefits are consumed, but not counted as part of
income, which decreases the saving rate;
 Higher capital gains tax revenue reduces disposable income and
hence saving
(these two items account for another quarter of the decline in the
personal saving rate.)

Consumer Durables purchases are a form of saving; including durables
would raise the level of the saving rates by about 3 percentage points, but
because there is no trend in recent decades it does not help explain the
decline in saving rates.
5
Tax Treatment of Capital Income
1.
Business firms earn the before-tax return to private investment; firms
then pay corporate taxes on these profits, at tc, the effective marginal
corporate tax rate;
2.
Suppliers of the capital receive the before-personal-tax return to savers,
and then pay personal taxes on their returns (interest, dividends, capital
gains) at rate tp, the marginal effective personal income tax rate on
capital income
Thus a “wedge”, or distortion, is created between the before-tax return to
private businesses on their investment and the after-tax returns to savers
supplying the capital for the investment
The tax wedge = tc + tp
The “excess burden” of tax goes up with square of tax rate
t c + tp
6
Long-Term Effects of Taxes on Saving and Investment
(real future value of $1 invested earning 10% before tax when taxed at 0% and 40%)
year 1
year 20
(1)
no capital tax
1.10
6.73
(2)
capital income tax
tc + tp = 40%
1.06
3.21
96%
48%
(2)/(1)
Tax rate on capital income = tc + tp
So, while a “small” capital income tax may not be too distortionary between
consumption today and consumption next year, it is very distortionary between
consumption today and consumption many years in the future.
7
It Is Difficult to Measure Capital Income
1.
Measuring capital gains and losses, changes in asset values, for
non-publicly-traded assets is difficult
2.
Measuring depreciation, decline in the value of an asset due to
wear-and-tear and obsolescence, for assets not sold on “used”
market is difficult; “lemons” problem
3.
Measuring real income, separating out real gains (or losses) from
inflationary gains or losses, is difficult
4.
Timing issues

Very difficult to measure real effective tax rates and to prevent
financial engineering, e.g. of equity into debt
8
Single and Double Taxation of Saving
1. Ordinary saving: taxed twice under the income tax

Earn income = $1, pay tax t, save $(1-t)

Saving earns return r, is taxed at t, your net of tax return is r(1-t)

Thus, you end up with (1-t)(1+r(1-t))
2. IRAs,




401(k)s: saving taxed once, on withdrawal
Earn income = $1, save and deduct it, so saving = $1
Saving earns return r, no tax while builds up
On withdrawal pay tax t (can be higher or lower than tax rate when
contributed)
Thus, you keep (1+r)(1-t)
3. Roth IRA: backloaded, saving not deductible, so no tax at withdrawal

Earn income = $1, pay tax t, saving = $(1-t)

Earns return r, no tax on withdrawal

So you keep (1+t)(1+r)
9
Tax Neutrality Toward Saving And Investment
1. Pure income tax: double taxation of saving
2. Pure consumption / consumed income tax: neutral
3. Add corporation income tax = third tax on saving
 Combination of tax rate, interest deductions on debt and PV of
depreciation allowances determines effect on investment (I)
 Tax income but allow true economic depreciation of equity financed
investment (no debt) => neutrality among types of investment.
4. Consumption tax: consumption (C) = income (Y) – investment (I)
 Neutral with respect to types of investment
 Neutral between investment / saving and consumption
 Immediate deduction of I called expensing (first year write-off)
10
Tax Neutrality Toward Saving And Investment
5.
Think of a football field
 Pure income tax level sideline-to-sideline, but saving (S) and
investment (I) running uphill
 Pure consumption tax – level sideline-to-sideline (among types of I)
and goalpost-to-goalpost (between S and C)
6. Debt raises the potential of negative effective tax rates (subsidies) on
investment
 The case for subsidizing investment
 The treatment of housing
7. Estate tax can be third or fourth tax on saving
8. Taxes and human capital investment
11
Human Capital Investment
Types:
Education (50%), On the job training (30%), Mobility, Health (20%)
Note:
Virtually all on the job training (ojt) and sizeable fraction of higher
education and some fraction of mobility and health costs are
foregone earnings, which ARE NOT TAXED.
Example: Early years of work and informal ojt (Heckman estimates 1/3 of
time).
So a $60,000 per year worker is actually “paid” $90,000 at annual
rate: $60,000 is cash and $30,000 is ojt
The in-kind ojt is not taxed. This is economically equivalent to
including it in income and then giving an immediate deduction
(expensing)
So a flat rate tax does not distort the bulk of human investment decisions.
However, a progressive rate structure does do so, as the human capital
investment raises earnings and drives people into higher tax brackets.
12
The Effects Of Taxes On Saving
1. Income taxes affect saving both by reducing the after-tax rate of return
received by savers and by reducing after tax incomes
2. By taxing the return (interest, dividends, capital gains) received by savers,
the tax makes it more expensive to save for future consumption, e.g.
during retirement. But the amount of saving theoretically could decrease
or increase in response to the tax (due to what economists call income
and substitution effects working in opposite directions)

Usually we think households will reduce the amount purchased of any
good whose price goes up

But, e.g. “target savers” would save more when the after-tax rate of
return declined
13
The Effects Of Taxes On Saving
3. The net effect of income taxes on saving is thus an empirical question

While there are a range of opinions based on various types of
evidence, the view of most economists is that saving is somewhat
responsive to the rate of return

Time series studies of aggregate consumption or saving for the U.S.
show that for each 10% increase in the rate of return (e.g. from 5% to
5.5%), saving increases between 1% and 5%

Evidence from the introduction or expansion of tax deferred saving
vehicles (e.g. IRAs, 401(k)s), which raise the rate of return to saving
show substantial effects

Every study of taxes and saving behavior, those that show larger,
modest, small or virtually no effects, is subject to a variety of critiques
14
The Effects Of Taxes On Saving
4. Even very modest (substitution) effects of capital income taxes imply very
large costs to society from the tax distortions between consumption and
saving (future consumption)

Nobel laureate Professor Robert Lucas, in his research and 2002
Presidential address to the American Economic Association,
concludes that removing the tax distortions to saving and investment
are by far the most important avenue for improving economic wellbeing of any potential public policy reform
15
Five Tests for Tax Reform
1. Will tax reform improve the performance of the economy?
2. Will tax reform affect the size of government?
3. Will a new federal tax structure affect federalism?
4. Will a new tax structure likely endure?
5. Over time, will tax reform contribute to a prosperous, stable
democracy?
16
Adam Smith’s Four Canons Of Taxation
1.
Equality: (Ability-to-pay) “…ought to contribute towards the support of the
government, as nearly as possible, in proportion to their respective abilities; that is, in
proportion to the revenue (income) which they respectively enjoy under the
protection of the state.”
2.
Certainty: “The tax which each individual is bound to pay ought to be certain, and
not arbitrary. The time of payment, the manner of payment, the quantity to be paid,
ought all to be clear and plain to the contributor, and to every other person.”
3.
Convenience in payment: “Every tax ought to be levied at the time, or in the
manner, in which it is most likely to be convenient for the contributor to pay it.”
4.
Economy in collection: “Every tax ought to be so contrived as both to take out and
to keep out of the pockets of the people as little as possible, over and above what it
brings into the public treasury of the state.”
17
Key Decisions For Design Of Tax System
1. Tax base(s): income, consumption, hybrid; people or
transactions
2. Tax rate(s): flat, progressive, levels
3. Unit(s) of account: family, individual, transactions
4. Time period(s) of account: transaction, annual, longerhorizon
18
Tax Reform
I.
Partial
Low rates
Simplification
Expanded tax deferred saving
Corporate tax integration
II.
Comprehensive
National retail sales tax
Value added tax
Consumed income tax
Flat tax
19
Alternative Ways To Tax Income And Consumption
(1)
Income = Consumption + Saving
or, Income – Saving = Consumption
(deductible saving method)
(2)
Income = Consumption + Investment
or, Labor Income + Capital Income = Consumption + Investment
or, Labor Income + (Capital Income – Investment) = Consumption
(business tax expensing method)
(3)
Excise, sales taxes
20
Consumption Taxes
(arguments for)
Equity
 People should be taxed on what they take out of the system, not
what they contribute.
 Taxing life-cycle income (leaving aside bequests and inheritances)
is equivalent to taxing consumption.
 Consumption may be a better measure of permanent income than
is current income.
 A consumption-based tax can be progressive.
Efficiency
 Eliminate distortion of saving decision and hence discrimination
against the patient vs. the impatient.
 Taxes all saving equivalently.
Simplicity
 A lot of the complexity of the tax code arises from capital income
taxation and avoidance.
21
Consumption Taxes
(arguments against)
Equity
 Income better measure of ability-to-pay (at least in an annual tax with
limited time horizons)
 Avoids differentiating capital and labor income
 Transactions based consumption taxes ineffective way of
differentiating by ability
 Transition unfair to elderly
Efficiency
 Intertemporal distortion not so severe; time horizons not so long
 In practice, may wind up subsidizing saving, investment (e.g., debt
finance and interest deductions unlikely to be limited in practice)
Simplicity
 Transition not simple
 Personal, business tax likely complex around saving, investment
definition
22
Tax Reform And The Transition To A New Tax System
1.
Tax reforms => capital gains and losses if the taxation of capital is
involved
2.
We not only must compare the new tax or tax system to the old one,
we also have to pay attention to transition issues
3.
Most major tax reforms involve transition rules
a. For a while, there are parallel tax systems:
i. The old one, phasing out and
ii. The new one, phasing in
b. Transitions impose costs, complexity
4.
Important to evaluate actual tax systems before, after and during
(transition), not just two theoretical systems
23
Conclusion

The U.S. saving rate is low and declining, although not nearly as much as
the NIPA measures;

The current tax system, on balance, is biased against saving;

Redressing this imbalance is very important and should be a major
component of tax reform;
 This is especially so, given the closely related issues in the evolution
of the public debt and social insurance transfers to the elderly;
24
Conclusion

The two routes to reform
1. Replace the current corporate and personal income taxes with an
integrated consumed income tax, e.g. with personal rates of 10%, 20%
and 30%, a corporate rate of 30%; phase out most deductions other
than charity and mortgage interest; include super-IRAs; simplify,
eliminate, combine other aspects of the code.
2. Replace the income taxes with more fundamental reform, such as the
Hall-Rabushka flat income tax or a national retail sales or value-added
tax;

The political economy of the sales and VAT taxes raise fundamental
concerns of federalism and potential growth of government
25