Lecture Twelve: Government Expenditure Analysis Part Two - Subsidies
Much of the growth in
government spending over the past several decades has taken the form of
subsidies for goods that are, or could be provided through the market mechanism.
Subsidies for education, food, child care, etc.
People would have bought them anyway (but perhaps not in the same
quantity).
How do people respond to
these subsidies? This depends in
part on what type of subsidy is used.
Fixed-quantity
subsidy – the government makes a certain quantity of a good available to a
consumer at no cost or at a cost below the market price.
The quantity of the good being subsidized is beyond the control of the
consumer – the government determines the quantity of the good made available.
(This is a form of in-kind subsidy, so called because the subsidy
is linked to the consumption of a particular good).
Example:
the government may provide food stamps that a consumer can use to
purchase $1,500 worth of food, but if more than $1,500 worth of food is desired,
the consumer must pay for the additional amount at the full market price.
Here the subsidy applies only to a given quantity, $1,500 worth of food.
Another example:
public schools make available a certain quantity or quality of schooling,
and if more is desired, it must be paid for by the consumer.
That is, if parents are not satisfied with the public school, they can
send their kids to a private school at their own expense – or pay for
additional tutoring, etc.
Fixed-quantity subsidies
can have various effects on the consumption and well-being of recipients,
depending on the size of the subsidy, the good being subsidized, and who pays
for the subsidy.
Reduced Private Purchases
One major impact of a
fixed-quantity subsidy is that is causes a reduction in out-of-pocket
expenditures on the subsidized good. Such
a reaction on the part of consumers is intuitively obvious because if the
government provides a good, consumers will need to purchase less on their own.
There are also situations
in which the consumers reduce their out-of-pocket expenditures by an amount
equal to the quantity provided by the government, so that their total
consumption does not increase.
GRAPH: Fixed Quantity Subsidy - Reduction in Private Purchases
Analysis:
Consumer budget line relating the subsidized good (food) to other goods.
The budget line reflects
the consumer’s income of $1,000 and the market price of food, $10 per unit.
The government gives the
consumer 30 units worth of food for free.
Note:
if the government had given the consumer a cash transfer of $300, the
cost to the government of the food subsidy, his budget line would be somewhat
different.
The effect of this
fixed-quantity subsidy on consumption opportunities is the same as a cash
transfer, except that the dotted portion of the budget line is not available to
the consumer.
This does assume that
resale is not allowed – otherwise the consumer could sell some food for $10
per unit.
The consumer’s exact
response depends on how the budget line is affected and on his preferences
concerning food and other goods. If
we assume that food and other goods are normal goods, then we know that after
receiving the subsidy he will choose a point involving more consumption of both
(he consumes 50 units of food and $800 worth of other goods.)
Recall that at the original equilibrium the consumer purchases 40 units
of food; after the subsidy, his consumption of food has increased by only 10
units, even though the government provides 30 units of food.
His private food purchases have fallen from 40 to 20 in response
to the subsidy, but total consumption has risen to 50.
Note also that the subsidy
has allowed the consumption of other goods to increase.
Before the subsidy, he spent $400 on food and $600 on other things; after
the subsidy he spends $200 of his own income on food (and receives a $300
subsidy) and has $800 left for other goods and services.
A reduction
in private
purchases should be expected with a fixed-quantity subsidy.
In addition, as long as the
quantity provided by government (30) is less than the consumer would purchase if
given the subsidy in the form of cash (50 units would be consumed with an
unrestricted cash transfer), this type of subsidy is equivalent in its effects
to a cash transfer
Taxes To Finance the Subsidy
But now we have to consider
the taxes to finance the subsidy. There
are two cases:
First,
someone other than the subsidy recipient may pay the required taxes.
In that case, our previous analysis is complete, at least insofar as the
effect on the recipient is concerned.
Second,
the recipient of the subsidy may also pay taxes to finance the subsidy, just as
many families pay taxes to finance the schools their children attend.
In this case, it is worthwhile to consider the combined effects of the
tax and subsidy on the consumer’s choice and well-being.
Assume that the tax is
equal to the subsidy ($300 in our case). This
would not normally be true for all consumers of course, on average the tax must
equal the subsidy (if a tax is used to finance the program).
Some would pay less, others would pay more than they receive in the
subsidy. But let’s assume each
pays what they receive to start wit
In this case the budget
curve would shift in when the consumer pays the tax, then shift back out again
when the subsidy is received. The
combined tax-expenditure policy leaves the budget line completely unchanged,
except for making it impossible to consume along the dotted potion of the
original budget line. If the
consumer would not have chosen a point on that portion of the budget line
anyway, he would have ended up purchasing the same quantities of food and other
goods.
These conclusions suggest
that many government expenditure programs may have little or no effect on the
allocation of resources.
However:
we have made a lot of assumptions. When
these assumptions are changed, the outcomes change.
Allocative and distributive effects of a fixed-quantity subsidy might
include:
1.
Over-consumption:
If the consumer receives the subsidy and pays no taxes – this analysis
indicates that a cash transfer and a fixed quantity subsidy are not different.
But there are times when a fixed-quantity subsidy will increase
consumption by more than a cash transfer. This
happens when the quantity provided by the government is greater than the
consumer would purchase if he had cash rather than the in-kind subsidy.
If he is given cash equal to more than 40 units of food (which he is
buying before the subsidy) he might prefer to use some of that cash on other
goods and would be better off with the cash instead of the food (from his point
of view anyway). He over-consumes
the subsidized good.
2.
Under-consumption:
It is often taken for granted that a fixed-quantity subsidy will increase
consumption, but we have seen that in some cases it may lead to the same level
of consumption as a cash transfer does. In
addition, there are cases in which this type of subsidy will actually reduce
consumption! This can occur when
it is very costly, or impossible, for the consumer to supplement the quantity of
the good provided by government.
Example:
Suppose the subsidized good is housing – and the government offers a
family a two-bedroom apartment at no cost.
The family may prefer a three-bedroom apartment and might be willing to
pay the difference in cost between a two-bedroom and a three-bedroom apartment
to obtain a larger apartment. The
way the program is administered, however, this option is not available.
The family cannot accept the government two-bedroom apartment and, by
paying the cost of an extra bedroom, convert it into a three-bedroom apartment.
Instead they must either accept the two-bedroom apartment of forgo the
subsidy altogether and pay the entire cost of housing themselves.
It is quite possible that the family will choose the two-bedroom
apartment when they government foots the bill rather than the three-bedroom
apartment they would have chosen in the absence of the subsidy.
Housing
is costly and “lumpy” – not easily divided into small units.
Therefore it makes it difficult, if not impossible, to supplement.
Two
studies offer evidence:
John
Kraft and Edgar Olsen studied a sample of public housing tenants and estimated
that 49 percent of the families were consuming less housing than if they had
been given the subsidy in cash.
Sam Peltzman studied the effects of state-supported colleges and universities on the consumption of higher education. He found that the expenditures per student would be higher in the absence of subsidies to higher education, which supports our contention that some fixed quantity subsidies result in under-consumption. Some students went to in-state public schools when they otherwise would have gone to out of state private schools, for example. Peltzman also found that more students attended college as a result of the subsidies. Thus, some students consume less schooling and others (those who would not have attended college without the subsidy) consume more as a result of state support of institutions of higher education.
3. Redistribution of Income: We have seen that the fixed-quantity subsidies can increase, reduce, or have no effect on the consumption of the subsidized good. The exact outcome probably varies widely from one subsidy to another and from one consumer to another. It should be pointed out that fixed-quantity subsidies often serve to redistribute income to certain groups.
For example, food stamps, housing subsidies, and Medicaid subsidies are concentrated exclusively on low-income household and therefore act to increase their real incomes. In addition, even though public schools and social security (subsidized retirement pensions) are more widely distributed, when account is taken of the taxes that finance them, we find that many families gain on balance, while other families lose; income is effectively redistributed. In general, it is plausible to suppose that the way such subsidies serve to redistribute income may be more important in an overall evaluation than their possibly minor impacts on consumption patterns.
Excise Subsidy
This
is a fundamentally different form of subsidy – the government pays part of the
per unit price of a good but allows the quantity of the good to be determined by
consumer purchases.
Example:
suppose the government decided to pay the consumer $5 for each unit of
housing consumed. The quantity and,
hence, the cost to the government depend on the level of consumer purchases and
so are not fixed by the nature of the policy.
This is an excise subsidy.
Other
examples: some welfare programs
operate as excise subsidies, but maybe the most common examples are found in
special provisions of the tax laws – like we have already discussed.
There
are two types of excise subsidies:
Ad
valorem excise
subsidies:
the government pays a certain percentage of the per unit cost of some
good or, what amounts to the same thing, a specific percentage of the
consumer’s total expenditures on the good.
Per
unit excise subsidies:
the government pays a certain amount for each unit of the good consumed,
as in the housing subsidy mentioned earlier.
The
allocative and distributive effects of both types of excise subsidies are nearly
identical. So let’s look at the
per unit excise subsidy.
Allocative
Effects: Market Perspective
Industry
Subsidy:
Let’s assume that the food industry is to be subsidized and that
the industry is a constant cost industry (to make our lives easier) . . . and
price competition among firms.
GRAPH:
Allocative Effects of An Excise Subsidy
The
initial price and quantity are $10 and Q1.
Now suppose that an excise subsidy of $4 per unit of food is to be
extended to the industry.
The subsidy could be paid either to the firms or to the consumers; we
will initially assume that it is paid to the firms.
This
reduces the firm’s production costs by $4.
Thus, the effect can be shown as a downward shift in the supply curve.
The industry has an incentive to expand output, and that leads to a lower
price for consumers.
The
ultimate effect of this subsidy, even though it is paid to the firms, is to
reduce the price to consumers by $4, the amount of the per unit subsidy.
(Note: if
this is an increasing cost industry, the price to consumers will not fall by
exactly the amount of the subsidy).
As
a result, consumers are confronted with a $6 price, and at the lower price
consumption increases to Q2.
With
an excise subsidy, consumption and output increase.
However,
we should remember that even though the supply curve shifts by the amount of the
subsidy, it is clear that the subsidy does not reduce the true cost of
production. Subsidies
do not reduce the marginal social cost of production – (given by the original
supply curve).
They just reduce the net costs to participants in the subsidized market
by having someone else (the taxpayers) bear part of those costs.
Consumer
Subsidy:
An excise subsidy can also be granted directly to consumers rather
than to firms.
That, in fact, is the way various tax subsidies operate.
Suppose that consumers receive $4 from the government for each unit they
purchase; the firms are not subsidized.
The subsidy then increases the per unit price that consumers are willing
to pay to firms.
So
consumers are now willing to pay $14 for the good at Q1.
In other words, the demand curve will shift up by $4. Again,
producers end up receiving $10 per unit – just as they did when the subsidy
was paid to them.
Also consumers are now paying a net price of only $6 – the remaining $4
is the subsidy and output has again moved to Q2.
Thus
we reach the surprising conclusion that an excise subsidy has the same effect,
regardless of whether consumers or producers are subsidized.
IN both cases the price of $10 is received by producers and a price of $6
is paid by consumers!
So we can analyze the subsidy either as an upward shift in the demand
curve or as a downward shift in the supply curve.
Over-consumption:
Note that the expansion of output represents over-consumption by the
consumers. The
only reason the consumers purchase the additional output is because someone else
is bearing part of the cost.
Is
there an efficiency or welfare loss then?
Would consumers, without the subsidy, prefer resources used elsewhere?
The subsidy results in an output level where the marginal benefit to
consumers is less than the marginal social cost of production.
Allocative
Effects: Individual Perspective
Greater
insight of allocative effects from an individual perspective.
The excise subsidy lowers the price of food, so the budget line rotates
and becomes flatter.
GRAPH:
Allocative Effects of an Excise Subsidy, (Individual
The
new equilibrium represents over-consumption:
the artificially low price encourages consumers to purchase more food and
less of other goods, and this is inefficient. This inefficiency can be demonstrated by assuming that the
consumer is given the subsidy in cash rather than as a subsidy that lowers the
price of food.
The
cash transfer would produce a parallel movement of the budget line.
This means that the consumer could still choose the consumption he wanted
if given the subsidy in the form of cash. The
consumer might prefer to purchase less food and more of other goods.
The consumer would be better off with an unrestricted cash transfer.
Therefore, compared to cash, the excise subsidy process a welfare cost:
the consumer is on U2 with the excise subsidy but can reach U3 by
consuming a different combination of goods of the same total cost with a cash
transfer.
Taxes: If the consumer doesn’t pay the tax or does pay the tax to
finance the subsidy – it doesn’t matter – the excise subsidy produces a
welfare cost by artificially stimulating food consumption in both cases.
The only difference is that the welfare cost associated with the subsidy
when others pay the taxes reflects the fact that the consumers would be better
off consuming less food with a cash transfer (could reach a higher U curve).
When the consumers pay the taxes themselves, the welfare cost reflects
the fact that consumers would be better off consuming less food without any tax
or subsidy. The nature of the
welfare cost is the same; an artificially lower price stimulates too much
consumption.
Distributive
Effects
Who
benefits and who loses from a subsidy depend on the exact type and size of
subsidy (as well as the distribution of the tax burden).
Where
the benefits from a particular subsidy accrue depends on the reaction of the
market affected by the subsidy. In
our earlier analysis we assumed a constant cost industry – so the $4 subsidy
went to the consumer as either a direct subsidy or lower price ($10 to $6).
But with an upward-sloping supply curve, this is not the case.
Part of the subsidy would go to the supplier.
For
example, if the subsidy does, on balance, increase consumption, this reflects an
effective increase in demand. Coupled
with an upward-sloping supply curve, a fixed-quantity subsidy that
increases demand will increase price. This relationship may explain some of the support by
educators (such as teachers’ unions and
colleges
and universities) for subsidies to education or by construction unions for
housing subsidies.
Or
the if the subsidy is given directly to the industry, the same result occurs –
with an upward-sloping supply curve.
GRAPHS
Taxation
and the Analysis of Expenditures
In
evaluating the efficiency effects of government spending, it should be kept in
mind that the fundamental issue is whether the spending program together with
its financing method leads to a more or less efficient resource allocation.
Normally, taxes will provide the funds that are spent, and we need to
consider somewhat more carefully how taxation is integrated into the analysis of
an expenditure program.
Central
point: every dollar the
government spends imposes a cost on taxpayers of more than a dollar.
There are several reasons for this.
1.
Administrative costs associated with collected tax revenues.
2.
Compliance costs borne by taxpayers – the time and resources required
to comply with the tax laws.
3.
Probably more important – resource allocation distortions – due to
less efficiency, these costs are very important but often over-looked.
These
costs (and possibly others) can be combined and referred to as the
marginal
welfare cost of taxation. There
meaning is, as already stated, that each dollar spent by the government has a
cost of greater than $1 on taxpayers.
For
example, if marginal welfare cost is 25% of revenue, than each dollar spent
costs the public $1.25 -- $1 of which is the direct cost of the revenue and the
other $0.25 of which is the marginal welfare cost.
These
costs have important implications for evaluating the efficiency of expenditure
policies. We have basically ignored
them in our analysis – thereby, for example, our analysis makes consumers
better off than they actually are (since we have ignored a cost).