Topic 7: Government Expenditure and Tax Policies


In the final Topic of in this Lesson we address the problem of how governments should spend tax payers' money and how they should acquire that tax revenue.

Government Expenditure Policy

There are three broad reasons why governments need to raise and spend tax revenue. First, a legislative process has to be paid for through appropriate compensation for the work of elected politicians and the employed workers that are needed to help them determine and carry out decisions that represent, as best as can be established, the will of the community. These decisions are necessary to ensure that an appropriate institutional framework is constructed and maintained within which private sector activities can efficiently function. An essential component of this expenditure pays for a legal structure which ensures that individual citizens, and organizations they create, adhere to rules necessary for proper functioning of these institutions. The second broad reason is the existence of externalities that arise because individuals, in many areas, do not receive and pay some of the social benefits and costs of their actions. Funds have to be raised and spent to bring, as far as possible, these individual costs and benefits into line with their social counterparts. A third reason is that, even if social institutions function perfectly, the community as a whole may be unhappy with the distribution of wealth and income among its citizens and may want to redistribute income from one part of the community to another.

An obvious requirement for government taxes and expenditure is the maintenance of a police force. While private police forces can ensure that the individual rights of the people that hire them are maintained, their focus is entirely on the interests of those individuals and only indirectly on, and possibly even against, the interests of the community at large. Another classic role for government is the provision of fire protection of property---were fire protection provided by a private market, individuals whose property must automatically be protected by fire fighters as part of their protection of a neighboring property will have an incentive to free ride off their neighbor's fire-protection expenditure. But the best example of a complete externality is the provision of national defence. Everyone in the community benefits from the protection provided by the national armed forces, completely independently of the amount any particular individual spends on the provision of those services. And the armed services must be entirely controlled by a representative government to avoid any possibility of a military dictatorship.

While government-imposed regulations can eliminate some externalities, the best practice in a free market economy is for the government to impose a tax or subsidy on the provision of the services to which the externality is attached. The standard example is toxic chemicals emitted in the process of producing an otherwise valuable commodity. The authorities can simply impose a tax on the output of that commodity sufficient to reduce that output to the point where the marginal social benefit from another unit of production---normally the market price---just equals the marginal social cost of that unit, where the marginal social cost is the sum of the marginal private cost to the producers plus the social cost of the damage resulting from its production.

Another externality arises with respect to the provision of roads. For the private sector to produce and manage a road system, the firms constructing and managing roads would have to obtain ownership of the land on which the roads are to be constructed and then, once a road is built, be able to impose tolls to earn revenues sufficient to cover its cost. It turns out that only in the case of major roads can the revenues from such tolls sufficiently exceed the costs of collecting them. And the acquisition of rights to build roads requires, in many cases, the legal powers that can only be invested in government. It is thus not surprising that almost all roads are built and owned by the community at large through the actions of government.

Since all members of the community benefit by having the people they have to interact with able to read and write and in possession of some minimum degree of knowledge about a wide range of issues, the acquisition of education by every individual yields external benefits to other people. Were the provision elementary and secondary education arranged and paid for entirely by the individuals acquiring it, those individuals would have no incentive to take into account these benefits to others beyond themselves. The common way of dealing with these externalities is for the government to provide elementary and secondary schools and force all young people to attend them. An alternative is for the government to provide every school-age individual with a voucher that must be spent to obtain attendance at a privately owned and operated school. This has the advantage that the private schools' survival will depend upon them offering the education that the students' parents want, forcing the school teachers and administrators to maximize the parents' utility rather than simply their own. Of course, since not all parents will have the understanding and ability to impose the necessary conditions on the private educators, the government most surely will have to impose regulations which would most likely include independently set and marked common examinations which all students would be required to take. There are a number of problems with this alternative approach. First, it may allow religious and ethnic groups to induce the incorporation of socially divisive elements into the education of certain groups of children. Second, it may lead to a hierarchical ordering of schools by student ability so that students of poor ability may have no opportunity to be educated along with those of greater ability---this would follow from the fact that many parents will try to get their children into the most educationally rigorous environment possible. Nevertheless, of course, it may be viewed by some educational experts as more effective to educate students in conjunction with others of similar rather than differing levels of academic ability. Finally, it might be argued that any efficiency and motivational gains that could be attained by a system of government-voucher-financed private schools could also be obtained within a purely public school system by simply imposing independently set and graded common examinations within that system---being forced to get one's students to a level where they do better on a common exam may be a more appropriate motivational tool than the profit motive.

Health care is a government regulatory nightmare. Setting aside issues involved in disease prevention, it turns out that in a free-market health care system in which government is not involved there are virtually no externalities. Those paying for medical advice and treatment from doctors and for hospital care impose no significant external costs on other individuals or provide them with significant external benefits. A free-market system will thus work wonderfully except for one problem: Really poor people will have insufficient income to afford treatment for even minor medical problems and, on frequent occasions, middle and even some upper income individuals faced with serious medical conditions will exhaust their wealth before death claims them. These problems for middle and upper individuals can often be avoided by purchasing medical insurance although, in an unregulated market, this would be impossible when specific future medical conditions can be identified in advance. And, of course, the really poor may not be able to afford medical insurance. As countries became richer and medical care became more effective, and costly, people became unwilling to watch others die for loss of treatment---especially given an inability to suppress the thought that it could happen to them. An attempt by the government to require that everyone buy medial insurance encounters the problem that very poor people and rich people with known or highly probable medical conditions will not be able to afford it, and private insurance companies cannot provide insurance at less than its cost. It is thus inevitable that governments are required to pay for the provision of some minimal level of medical care to anyone who cannot afford to pay their own way. In some countries, the government deals with these problems by financing medical care for everyone in the community, with the costs being paid out of taxes. Immediately, externalities arise. The authorities have to regulate the fees charged by doctors and hospitals because those individuals and institutions have every incentive to set those fees as high as possible. Then what is to stop doctors from scheduling unnecessary future appointments for their patients in order to increase their income? And what is to stop patients from arranging unnecessary visits to doctors? (The British system of making every person have a doctor and then paying doctors a salary for each patient on their list suffers from the fact that it weakens the incentive for doctors to provide really good service.) Then there is the willingness of many families to spend millions of dollars of public money to keep grandma alive for another week---the decision to pull the plug can no longer be an entirely private decision, raising ethical issues that have to be agreed upon by the community as a whole. Since there are no natural limits on monetary costs, the rationing mechanism for such medical care systems is increased waiting times which, it always turns out, will be surmounted by those having important personal within-system contacts or political influence. Important pressures thereby arise to permit individuals to purchase medical care directly or by taking out private insurance, thereby creating a two-tier system. Given the pressure on politicians to finance additional programs elsewhere in the economy, they have a political incentive to scrimp on the provision of public medical care, thereby further increasing waiting times and expanding private-tier provision and reducing the quantity and quality of service to the poorer members of the community. How should these issues be best dealt with? No one really knows!

Tax Policy

We now turn to the question as to the appropriate ways governments can raise the tax revenue to finance their expenditures. Policies of taxing the outputs of particular products to appropriately reduce associated negative externalities will raise revenue for subsequent expenditure. An obvious use of this revenue would be to appropriately subsidize any products whose outputs generate positive externalities. Any balance of such revenues can go to finance other government activities.

In general it would seem reasonable to impose the taxes required to finance government policies on the individuals who benefit from such policies. For example, taxes on gasoline and diesel fuel can essentially charge users of roads for the costs of their construction and maintenance. Property taxes can also be used to charge beneficiaries of sidewalks that service those properties and lawns and trees that beautify the roadsides. Property taxes might also be a good way to finance government expenditures on fire protection. While police protection also benefits property owners, it also services the community more generally and would seem to be best paid for by more general widely applied tax policies. The same applies to financing the public school system. Medicare could be reasonably financed by flat premiums charged to all individuals covered, although adjustments would have to be made, or other methods used to enable the very poor segment of the community to cope with these costs.

There are essentially two main types of broad-based taxes that can be used to finance the wide range of government expenditures that benefit large and diverse segments of the community. These are income taxes and either generalized sales or value-added taxes. Income taxes generally tax away some fraction of each persons income. They can be designed to garner the same fraction of everyone's income or, alternatively, to take a fraction of the income that increases with the person's income level. Generalized sales taxes or equivalent value-added taxes, normally denoted by the respective terms GST and VAT, tax the values of the outputs of a wide range of goods produced and sold in the economy. Generalized sales taxes are imposed on the selling prices of final products, with all intermediate products entering into their production being exempt, while value-added taxes are imposed on the value added at each stage of the final good's production process with each firm involved in that process being taxed according to the value it contributes to the final selling price. These two taxes differ primarily in the method by which they are collected with the total tax taken as a percentage of the final selling price charged consumers being essentially the same.

To analyse the effects on the economy of these alternative types of broad-based taxes, it is necessary to think in terms of the underlying process of economic growth. The full-employment level of output can be expressed in terms of the following proportional relationship between output and the economy's capital stock, broadly defined to include human capital, technology, and knowledge as well as the usual forms of physical capital.

             1.    X  =  (m  −  δ) K 

where  X  is the level of output,  m  is the output flow per unit of aggregate capital stock,  δ  is the rate of depreciation of capital and  K  is the aggregate capital stock. The level of  m  will depend on how efficiently aggregate capital holdings are divided into the various types of capital and on how efficiently those capital resources are used to produce output---it will achieve its maximum value under conditions of perfect efficiency. One condition of efficiency will be the extent to which the governmental process establishes an appropriate legal structure, correctly regulates the private sector and produces publicly those goods and services which government can produce more effectively than can the private sector. Suppose that a resulting level of  m  becomes established. Given the rate of depreciation, this implies a specific level of aggregate output. Aggregate income will typically differ from aggregate output to some degree as a result of the fact that some of the domestically employed capital stock may be owned by foreigners and domestic residents may own some capital stock employed in the rest of the world.

It is reasonable to assume that domestic residents will take the future into account when making each period's aggregate consumption, saving and investment conditions, maximizing a utility function that can be represented in the following inter-temporally additive form

             2.    U  =  U(Ct )  +  [1/(1 + ρ)] U(Ct+1)  +  [1/(1 + ρ)] 2 U(Ct+2)  +  [1/(1 + ρ)] 3 U(Ct+3)  +  [1/(1 + ρ)] 4 U(Ct+4)  +  ...............................

where  U  is the level of utility in the beginning period, U(Ct+i )  is the utility received in year  t+i  and  ρ  is the rate of time preference. Equation 2 gives the present value of all future years' utility from consumption. While this particular form of the inter-temporal utility function is adopted for simplicity, it nevertheless contains all the necessary attributes for the discussion that follows. Given the interest rate determined in the world at large, individuals will allocate consumption through time so that there is zero gain from shifting a unit of consumption between any two adjacent periods. This implies that the marginal utility of the last unit of consumption in any period be equal to the present value of the utility that would be received by shifting that unit of consumption to the next period.

             3.    U′(Ct ) dCt  =  [1/(1 + ρ)] U′(Ct+1) (1 + r) dCt

where  U′(Ct )  is the increase in utility that would result from increasing consumption in that year by  dCt.  The term on the left side of the equality thus gives the change in current utility associated with a tiny change in consumption in that year. The term  (1 + r) dCt  gives the increase in next-period consumption that could be obtained by investing the amount  dCt  for one year at the interest rate  r  and then consuming the result. The collection of terms  U′(Ct+1) (1 + r) dCt  gives the amount of utility that will thereby be obtained next year, and the term  1/(1 + ρ)  discounts that future utility back to the current year.

Equation 3 can be expanded by multiplying the  (1 + r)  term by  (1 - τ)  where  τ  is an implicit tax, which could be positive or negative, on savings---that is, on future consumption---imposed indirectly by government policy. This tax reduces (or if negative, increases) the amount of next-period consumption that can be obtained from foregoing  dCt  units of current consumption.

             4.    U′(Ct ) dCt  =  [1/(1 + ρ)] U′(Ct+1) (1 + r) (1 − τ) dCt .

Dividing both sides of the above equation by  dCt  and rearranging the terms yields

             5.    U′(Ct ) / U′(Ct+1 )  =   (1 + r) (1 − τ) / (1 + ρ) .

If there is diminishing marginal utility in each period, the ratio of marginal utilities to the left of the equality is directly related to the rate of growth of consumption---if consumption increases between this year and next, the marginal utility of consumption will be lower next year that it is this year. The marginal utility of consumption this year divided by the marginal utility of consumption next year will thus be positively related to the rate of income growth. Moreover, if  r ,  τ  and  ρ  are constant through time, the growth rate of income will be constant through time.

Suppose now that the government finances a broad range of expenditures by the imposition of a tax that takes a fixed percentage of everyone's income independently of the level of that income. Since income is taxed independently of whether it is spent on consumption or saved, this tax will not distort the communities choice between present and future consumption---apart from other possible distortions,  τ  will be zero and the rate of growth of income in the economy will be unaffected. Of course, the tax will affect peoples' choices between work and leisure but the equivalent free provision of goods and services by the government will tend to offset this effect. Alternatively, suppose that the government taxes away a percentage of income that increases with the level of the person's income. This will mean that future income will be taxed more heavily than current income and people will have less incentive to save--- τ  will become positive, reducing the growth rate of the economy. Everyone in the bottom half of the income distribution will have an incentive to vote for this progressive income tax and, if the rich care about the suffering of the poor, at least some people in the top half of the income distribution will also vote for it.

An alternative approach is to impose a flat-rate GST or VAT on all goods sold in the economy. This is equivalent to a tax on income except for the fact that savings channeled into government bonds will not be taxed and savings channeled into into shares in private corporations may not be taxed unless the tax is imposed on the production of all capital goods and services purchased by corporations. To the extent that savings are not taxed, there will be an increased incentive to accumulate capital, resulting in a reduction in  τ  and an increase in the economy's growth rate. It would be possible, though a very difficult undertaking, to combine this tax with an income tax that is just sufficient progressive to offset any positive effect of this tax on economic growth.

In this respect, one might argue that the main reason to vary income tax rates with income is to avoid taxing away the incomes of the very poor members of the community, rather than to tax the very rich relative to the middle class or the richer individuals in both of these groups relative to those who are less rich. A way to accomplish this would be to levy a proportional income tax but allow the very poor to claim a negative tax---that is a topping up of their income. While this would give the poorest in the community some incentive to work less, this negative work-incentive would be no bigger than would result from giving poor people welfare benefits, in excess of any income earned, sufficient to bring their incomes to some minimal level. And a negative income tax would probably also be cheaper to administer than a social welfare system.

Now what about corporate income taxes? To the extent that these taxes fall on income that is paid out in dividends, they involve double taxation when combined with taxes falling on income. This will create an incentive to consume rather than save, increasing  τ  and reducing the growth rate. Taxation of dividends, or of people's income in general, will also create an incentive for corporations to reinvest earnings rather than pay them out in dividends, thereby channeling the income into future capital gains. Under these conditions, social efficiency requires that capital gains be taxed appropriately. One way around this problem would be to tax any income of corporations that is not paid out in dividends---this will avoid double-taxation of dividend income while at the same time taxing any income channeled into capital investment that will yield capital gains, making a separate a capital gains tax unnecessary. Another way would be to appropriately tax corporations on all their income while allowing people paying income tax to deduct from their income any dividends that have been received.

More generally, what is needed is an appropriate blend of goods and service or value added taxes, personal income taxes and possibly corporate income taxes as well, to produce an efficient overall tax structure, taking into account the community's income distribution objectives, that creates minimal distortion of people's choices between consumption and saving and, of course, also leads to the efficient mix of goods being produced.

Finally, it is time for our last test. Please think up your own answers before looking at the ones provided.

Question 1
Question 2

Choose Another Topic in the Lesson