Factor Pricing and  the Determination of Incomes


Wages and Income Differentials


Land and Rent

Income Distribution and Redistribution

Measuring Income Distribution

The Goal of Redistribution Policy

A "Just" Distribution of Income?

The markets studied to this point have been those in which finished goods and services have been produced and consumed. This topic will be concerned with the functioning of markets for productive factors, the inputs used to produce final goods and services. The main objective will be to show how the prices of productive inputs are determined and how this in turn determines the levels of income enjoyed by their owners. 

Labour is by far the most important factor of production, both in terms of the total quantities used in production relative to other inputs and also as a source of personal incomes. Most people derive some, if not all, their income from selling labour services. Only a minority derive significant earnings from selling producers the right to use land or capital. Even in the developed, industrial nations relatively few people own significant amounts of property.

The determination of wages and other factor earnings in a market economy is analytically identical to the determination of product prices. The task here is to identify the peculiarities of factor markets, the most basic of which is that producers have a demand for factor inputs because of what they can produce. The demand for factors of production is a derived demand. The other side of the market, on which the supply of productive factors is determined, is governed by the choices made by individuals with respect to the terms and conditions on which they are willing to make the services of their productive factors available to employers.

To Top


Beginning with the supply side of the labour market, it must be admitted that the concept of a "supply of labour" is abstract in the extreme. Labour services exist in a bewildering variety of types and qualities. The services provided by a brain surgeon and by an unskilled manual labourer are obviously very different. Even leaving aside the problem of "quality", it is difficult to measure the quantity of labour potentially available in an economy. How much labour is available must depend on the size of the population, on the proportion of that population which for reasons of age and inclination is willing to sell labour ("the labour force"), the conventions which determine the length of the working day, the number of working days per year, and the ages at which individuals enter or leave the labour force. There is also the difficult question of the "intensity" with which workers work. One hour of determined effort by a highly motivated person who is enjoying the activity may be several times more "productive" than an hour of labour performed by a person simply killing time on the job. These are all important considerations, some of which will be taken up later, but the first task is simply to identify the influences which will, in general, and given all these background influences, determine the quantity of labour services an individual will supply.

As in product markets, an increase in the price offered for labour would be expected to draw out a larger supply of it. It is consistent with common experience that the amount of labour individuals will offer varies with the wage rate offered. It is not difficult to imagine, even in the absence of a social welfare system, that there could be a wage rate below which some individuals simply would not bother to seek a job. When the rate rises above such a level, whatever it may be, people begin to supply labour, exchanging leisure time, freedom and independence, for the income which can be gained. If work and leisure are thought of as two alternative uses of time, a change in the remuneration for work will result in work being substituted for leisure. It is also possible, however, that when rates of pay reach high levels, the incomes of such fortunate individuals may become so large that they may choose to reduce the amount of work they do rather than increase it. This is analogous to the "income effect" of a price change in product markets. As income rises, the amount of leisure a person desires may also rise. Putting the two effects together means that once wage rates rise above some minimum level an individual will supply more and more labour when offered yet higher rates, but at some point may reverse this behaviour when the income effect overwhelms the substitution effect. Graphically the concept may be shown as a "backward sloping supply curve for labour".

Just as market supply in product markets could be determined by summing the quantities individual firms would supply at various prices, the supply of labour offered in a particular market is the sum of the amounts individual workers will offer at various wage rates. Because any negative effect an increase in wages might have on the amount of labour offered by some individuals in the market would likely be offset in most situations by other workers being attracted to the market by higher wages, it is usually assumed that the overall relationship between wage levels and the supply of labour in a market is positive — higher wages will be associated with more labour being supplied, lower wages with less.

What will the supply of labour look like from the perspective of a single employer in a particular labour market? Here again an analogy with product markets presents itself. The answer depends on whether or not the particular employer is a significant user of labour in the market as a whole. If a firm is a relatively small employer in a large labour market, it is unlikely to have any effect on the price of labour in that market. Consequently the firm can hire as many workers as it wants to at the going wage rate. Just as a perfectly competitive firm in a product market has no effect on the price of the product and can produce as large an output as it chooses without causing the product price to fall, so can a firm employing labour in a perfectly competitive situation simply take the existing wage rate as "given" and employ as many workers as it chooses to. 

A firm has a demand for labour because of the contribution that labour can make to the firm’s revenues. The latter is determined by two things: the quantity of additional output a unit of labour input can produce and the price the firm can sell that output for. Consider a simple example. Suppose a college student sets up a business mowing lawns during the summer break. The student manages the business and hires others to do the lawn mowing work. The question is, how much labour should be hired? Obviously this will depend on a number of things, including the demand for such services in the community where the business is located. Because it is not difficult to set up small businesses like this, it may be assumed that there will be plenty of competition. In fact, it may be reasonable to assume that conditions will approximate perfect competition in product markets. The quantity of labour employed by our student entrepreneur will be determined by the physical productivity of the workers, the revenue generated from their physical output, and the costs which must be incurred to hire them.

Output, measured in terms of the number of lawns mowed per day, depends on the number of workers hired. The more workers, the more output. But, as the work force increases, the increase in output is unlikely to change in the same proportion as the increase in the number of workers employed. If this seems familiar, it should. The reason has to do with the principle of diminishing marginal physical productivity. Because of this effect, total revenue increases as workers are added, but at a diminishing rate. If this enterprise was not operating in a perfectly competitive product market the rate at which total revenue increased would be even less than this, because then it would be affected not only by the diminishing marginal physical productivity of labour, but each additional unit of output would be sold at a lower price than the preceding one. 

The relevant values involved in such a situation are the average revenue product (total revenue divided by the number of workers) and this will decline as more labour is employed. The marginal revenue product (the change in total revenue divided by the change in the number of workers) will also decline. 

Now consider what the firm will have to pay its employees. Assuming that it is operating under conditions of perfect competition in the labour market as well as in the product market, the business will simply have to pay the "going rate" for the type of labour it requires. Suppose this is $60 per day. The decision rule for determining the optimum number of workers to hire under these conditions is intuitively appealing and similar to the decision rule whereby a firm determined its best level of output. In this case the firm should hire an additional worker if that worker would add more to revenues than to costs. In general, the point is to hire additional labour until the marginal revenue product (what the labour is producing for the employer) is just equal to marginal resource (in this case, labour) cost, the cost of hiring an additional unit of it.

Now the firm’s demand for labour can be defined. It is, in fact, identical to the marginal revenue product values. Not surprisingly, the firm’s demand for labour shows an inverse relationship to price: the lower the price of labour (the wage rate) the larger the quantity of labour demanded. If the demand for labour of all the firms in the market were to be summed the result would be the market demand for labour. Combine that with the supply of labour to the market and it is possible to determine the "going rate" of wages and the equilibrium quantity of labour demanded and supplied at that rate. 

Given the demand and supply conditions, the equilibrium price of labour in this market is determined and the equilibrium quantity of labour demanded by a typical firm hiring workers in that market  is also determined. If there are many such firms and no one of them has any significant influence on the market, the firm determines the quantity of the factor to employ by equating the going wage rate (its "marginal resource cost") with its marginal revenue products as shown here.

This equilibrium will change, of course, if anything happens to alter the supply or demand conditions in the market. An increase in the supply of labour, resulting perhaps from a change in social attitudes with respect to gender roles, would cause an increase in the supply of labour and (ceteris paribus) this would cause the equilibrium price to fall and more labour would be employed. A decrease in labour supply, perhaps as a result of a successful attempt to unionize workers in a particular market, would cause the equilibrium wage rate to rise and the quantity of labour demanded to fall.

It is also possible to infer from this analysis something of the general effect an "artificial" wage level would have on such a labour market. Suppose a law was enacted making it illegal for employers to pay a wage less than the equilibrium wage. This would result in an excess quantity of labour being offered relative to the quantity which would be demanded at that price as shown here. Similarly, it should be easy to see what the effect would be if a price was set below the equilibrium market rate. Would there now be a shortage or a surplus of labour in this market? (Policy  issue: Minimum wage legislation.)

The labour market considered here is the simplest of a number of possible (and more realistic) situations. It has been assumed that the typical firm is operating in a perfectly competitive labour market and a perfectly competitive product market. More realistic situations would include combinations of perfect and imperfect markets: a monopsonistic firm (the only buyer in a factor market) selling in a competitive product market, a monopsonist selling in a monopolised product market, etc. The effects of shifts in labour demand and supply in such markets may be different than in the perfectly competitive case, as may the impact of unions and minimum wage laws. Even so, the competitive case provides much of what is needed to explain how labour markets determine people’s incomes and also, as the next section will show, why such incomes are as unequal as they are.

To Top

Wages and Income Differentials

The discussion to this point has referred to "the" labour market , but there are in fact many quite separate labour markets in the real world economy. In some of these markets equilibrium wage rates are high, in others they are comparatively low. On the basis of the analysis so far it may be said that if the supply of labour is low and the demand high, wages will be high; if the supply is high and the demand low, wages will be low. But what determines these conditions?

So far as the demand side of labour markets is concerned, it has already been shown that two main influences are involved in the determination of the level of demand. One is the physical productivity of labour, the other the price of the product. The latter depends, of course, on the relationship between demand and supply in the product market. The former, the physical productivity of labour, depends upon a number of things. A particular worker’s health and physical condition will obviously be an important determinant of his or her physical productivity. So will the worker’s innate ability, as measured perhaps by intelligence, motor skills, and motivation. There is not much the worker can do about such natural endowments except adapt to them as best he or she can. Not everyone can have the kind of reflexes necessary to become a successful racing driver or the intelligence needed to break new ground in the field of cosmology or the grace needed to succeed as a diplomat. People fortunate enough to possess such endowments are likely to earn very high incomes indeed. The reason is that the supply of such talents is apparently very small relative to the demand for the services they can produce. This accounts for the astronomical incomes of great sports figures and other entertainers. Indeed, if the person’s talent is "unique" the supply of his or her services may be considered to be completely inelastic.

When supply is completely inelastic the wage rate (if it can be called that in this context) is determined solely by the demand for whatever extraordinary thing it is the supplier of such services possesses. The market for such unique abilities is, however, highly specialised and the person who can supply them may have few alternative employment opportunities. 

Imagine a popular (but otherwise unskilled) performer whose next best alternative employment is as a bus boy in a restaurant. In such a situation the real cost to society of having this person work as a performer is the cost of the foregone alternative, which in this case is the value of his services as a bus boy. The difference between that and what he earns as a high-paid performer is entirely attributable to the unique ability that makes him popular. If such ability is truly unique, it cannot be replicated. For reasons explained later when the pricing of land as a factor or production is discussed, whenever any factor of production is in fixed supply like this, anything it earns in excess of what is needed to make its services available is called "rent." 

The demand for unskilled workers is lower than the demand for more highly trained workers because their marginal productivities are different. Part of this difference may be attributable to differences in the physical productivity of skilled and unskilled workers, but more of it is likely to be the result of differences in the prices paid for the goods they are helping to produce. Physicians are highly paid because the health care services they provide command a high price and the supply of physicians is limited by the cost and other demands associated with the training they are required to have. Counter help in fast food outlets are paid low wages because very little training or experience is required (the supply of such labour is plentiful) and the prices of the products they are producing are low.

Many kinds of abilities can be learned. Most labour markets are differentiated on the basis of the kinds of skills and training required. A glance at the employment opportunities pages of the daily paper will show that, while some employers are seeking persons with certain natural abilities, most specify some level of education, training, experience or combination of these things, all of which can be acquired. There is an interesting controversy over whether such requirements are directly related to the skills needed to perform certain tasks or whether they are simply used to screen prospective workers to select for more general abilities. Any university degree, for example, may meet an employer’s requirement because it is not important what the applicant actually knows, only that he or she has whatever it takes to earn a university degree. 

Evidently acquiring skills is rewarding insofar as it permits higher earnings. But of course there is a corresponding cost. Acquiring skills through apprenticeship and similar kinds of on-the-job training means working for several years at low wages. Acquiring skills through formal education, as in the case of taking university degrees, is also expensive, not just in terms of fees, text-book expenses and the like but, more importantly, the years of income earning foregone. How great the net payoff from such "investment in human capital" is for the individual involved has been controversial, as has the estimated benefit to society at large. The evidence suggests, however, that it is positive in both cases.

The productivity of labour may depend not only on the abilities of workers and their motivation, but also on the materials and equipment they have to work with. Workers using good quality raw materials and assisted by effective machinery will be more productive than workers using low quality materials and primitive tools.

The difference between the most dissimilar characters, between a philosopher and a common street porter, for example, seems to arise not so much from nature, as from habit, custom and education.

—Adam Smith
To Top


Unless otherwise stated, the term capital will be used here in the sense of "produced goods for use in further production." This refers to the stock of machinery and equipment, structures and other real physical goods which is available to assist in the production of other goods. Although many difficulties stand in the way, this stock of capital can be measured, making it possible to track the growth of the capital resources available in a particular economy over time and to compare the quantity of capital in different economies. Although allowance must be made for differences in the quality of the capital stock, how up-to-date it is, for example, generally speaking the richer, more developed economies have much higher stocks of capital in relation to their supplies of other factors than do the poorer, less developed countries. The cause and effect relationship in this generalisation goes both ways. The more capital there is, the higher the productivity of labour and the higher per capita incomes can be. And the higher per capita incomes can be, the greater the ability to create and maintain a large capital stock.

One of the things that makes measuring the capital stock difficult is that it is constantly changing. This is due to at least two influences, obsolescence and depreciation. The higher the rate of technical change in an economy, the more quickly some of its capital stock becomes out of date. Consider, for example, how quickly computers become obsolete and are replaced by newer, more capable models. There is a close connection between technological change and the capital stock. For new technology to become effective, it usually must be embodied in capital. Nuclear energy technology, for example, would be of no practical importance unless it was built into a nuclear power station.

The other influence constantly at work on the capital stock is its physical deterioration in use. Machinery and equipment and buildings wear out as they are used. This means that some part of the capital stock is constantly in need of replacement or repair. This physical degradation of capital is called depreciation in economics. Note that accountants use the same term, but mean by it some arbitrarily defined rate at which the financial value of capital assets is reduced.

As noted at the beginning of the course, the act of creating capital, whether to replace worn out capital or to enlarge the capital stock, is known in economics as investing. The amount of investing that is possible depends on the amount of saving, which is defined as the act of abstaining from consumption. If some individuals or households in an economy have incomes in excess of what they choose to spend on consumption, the income not spent is called saving. It is also possible for firms and governments to save. Corporations, for example, often do not pay out all their net earnings to shareholders, choosing instead to retain some of these earnings for the company’s own use. Governments can also save by spending less during some accounting period, say a year, than they collect as tax revenues during the same period. But the bulk of savings in the modern industrial economies is done by individuals or households. The motive for saving is sometimes to finance an act of investment carried out by the savers themselves, most commonly to finance the purchase or construction of a home, but most investing is done by businesses who borrow funds from individuals or households to finance business investment.

Apart from saving to finance housing, the motives for saving are varied. The amount individuals can save usually depends on their stage in the life cycle. Young people seldom save because their incomes are low, but as they become better established their incomes rise and they become concerned about such things as putting money aside for their children's’ education or their own old age. 

The funds saved by individuals and households are typically accumulated in accounts held in various kinds of financial institutions, such as banks and trust companies, which then lend them to borrowers, or they are used to buy stocks and bonds. Unless they are hoarded as currency hidden away somewhere, savings consequently find their way into the hands of businesses which want them for the purpose of financing the acquisition or creation of real capital. 

Business borrowers of such funds must, of course, pay a price to have the use of them. This price is called the interest rate. Although there are a great many different interest rates in the real economy, varying with the credit-worthiness of particular borrowers, the time period of the loan, and the type of financial instrument used, these rates tend to move more or less in unison, giving some justification to the practice in economics of referring to "the" rate of interest to mean the whole set of interest rates in effect at some particular time.

Capital is nothing but the sum total of intermediate products which come into existence at the individual stages of the roundabout course of production.

—Eugene Von Böhm-Bawerk

The rate of interest may be thought of as being determined in a market much like any of the markets studied earlier. There is a supply of investable funds which arises from acts of saving and there is a demand for these funds from business firms who want to borrow them to finance the creation of new or replacement capital goods. The supply of funds offered is assumed to be determined by the price offered for their use, that is, by the rate of interest. To the extent suppliers of savings are affected by interest rates, more will be supplied at higher than at lower rates. The demand for funds may be assumed to show the reverse of this relationship. A larger quantity of funds will be demanded when interest rates are lower and a smaller quantity when interest rates are higher as shown here. Why this may be so requires some explanation.

Whether a firm will want to borrow funds to finance investment or not depends on the expectations of its owners or managers as to the likely profitability of such an investment. Whether it will pay the firm to invest in a new machine, for example, requires an estimate of the expected net earnings from that machine over its useful lifetime. This is likely to be a difficult estimate to make, for much will hinge on future sales prospects and other uncertainties. Experience, good judgement and luck may all enter into making good predictions of this kind. What is needed, however, is at least a guess as to how much the new machine will increase the firm’s earnings over time. If this can be expressed as a percentage annual rate of return on the investment, it can then be compared directly with the cost of borrowing the money needed to make the acquisition. The higher the interest rate the higher the expected return on the capital will have to be to justify undertaking the investment. Looking at all such investment decisions in the economy as a whole, high interest rates mean that relatively fewer such investments will be undertaken, lower interest rates make it likely that more will be. The quantity of investable funds demanded consequently varies inversely with the rate of interest. This is why the demand for investable funds displays the negative relationship between price (the interest rate) and the quantity of funds demanded.

How sensitive the demand for investable funds is to changes in the interest rate (as implied by the slope of the curve) is a matter of controversy and the reasons for this will be examined later. But note that a change in anything other than the rate of interest that affects investment demand will cause the overall demand to shift. For example, if business people become more optimistic about future business prospects, the demand for investable funds will be greater over the whole range of interest rates. Assuming that there was no offsetting shift in the supply of funds, this would cause the equilibrium interest rate, to rise. Here such a shift is shown graphically. Alternatively, if something happened to cause an increase in the supply of investable funds (ceteris paribus), this would cause the equilibrium rate of interest to fall. All these possibilities will be important when attention is given to the role of money and investment in determining the level of output in the economy as a whole later in the course. At this point what is important is to establish the connections between saving and the creation of real capital and to show how the rate of interest is determined in the market for investable funds.

Once established, the interest rate functions like any other price to ensure that the scarce supply is allocated to those who are best able to use it. In this case, investable funds are scarce and the market for them sees to it that they go to the highest bidders, the users who think they can make the most profitable (and hence productive) use of them.

No one supposes that the owner of urban land performs, qua owner, any function. He has a right of private taxation: that is all.

—R.H. Tawney
Land and Rent

So far it has been shown that the supplies of both labour and capital can be increased. The only factor of production which is in fixed supply is land. The natural resources of the planet are finite. This means that the price paid for the use of land is determined solely by the level of demand for it. As in the cases of labour and capital, the demand for land arises because of the value it can be used to produce. This value, as in the other two cases, is determined by the physical productivity of the resource and the price which can be obtained for each unit of physical output it produces, or what has already been defined as the factor’s marginal revenue product.

Certain choice building sites, such as land in the centre of Tokyo, for example, have a very high marginal revenue product. Land in remote rural regions has a relatively low marginal revenue product. Combine this difference in the potential earning power of these two kinds of land with the fact that the amount of land in downtown Tokyo is very scarce compared to the amount of remote rural land (even in Japan) and it is easy to see why the rate of return to owners of the former is certain to be much greater than to owners of the latter. What struck the first economists to study the pricing of land, however, was that since the supply of it is fixed, its price must be determined solely by the level of demand for it. Whether its price (rent) is high or low depends on whether the demand for it is high or low. Here we find a use for the concept of a perfectly inelastic supply curve introduced in an earlier topic.

This analysis, while simple, resolves one of the puzzles which perplexed policy makers and economists in England early in the 19th century. British manufacturers were complaining (then, as always) that their costs of production were being kept high because of the high wages they had to pay their workers. These high wages were caused, they believed, by the high cost of living which, for most workers of the time, was largely a matter of buying food. Basic food stuffs, such as bread, were expensive the manufacturers believed, because of the high price of grain (which they called corn). And the high price of grain they blamed on the exorbitant rents that landowners, the political adversaries of the manufacturing interests, charged farmers for the use of their land. David Ricardo refuted this analysis by showing that, as he put it, "corn is not high because a rent is paid, but a rent is paid because corn is high." In other words, as has just been illustrated, the amount paid for the use of land is determined entirely by the level of demand for the land. 

(The public policy implications of the economic analysis of rent were further developed in 19th Century North America byHenry George who proposed that all taxes be replaced by a "single tax" imposed on income in the form of rent. Such a tax, he argued would have no negative effect on production because rent was a payment made for the use of inputs -- land -- the value of which was determined, as Ricardo had shown, only by the demand for them.)

The term "rent" subsequently came to be used in economics to refer to that part of the earnings of any factor that is for some reason either completely or to some degree fixed in supply. To understand the implications of this it is necessary to recognise that the price paid for the use of a factor of production reflects two different influences. If a firm wants to use a factor of production currently employed in some other use, it must pay the owner of that factor enough to take it out of that alternative use and thereafter enough to prevent the owner from transferring it to some other use. This payment is called a "transfer earning" of the factor owner. In a well-functioning market, the transfer earnings of a factor will be equal to the value it produces in its best use. 

Transfer earnings may be all the earnings owners of certain kinds of resources receive. An unskilled worker looking for work in a crowded job market is unlikely to get a job paying anything more than the going rate for unskilled labour. Contrast this with the case of a person with some unique ability, as discussed earlier.Such a person may command a wage in excess of what he or she could earn in the next best alternative employment. The total remuneration obtained in such a situation would consist of transfer earnings plus this additional amount attributable to some uniqueness. The latter element is analogous to the fixed supply characteristic of land resources and for this reason the return to a resource owner in excess of the transfer earning is called an "economic rent". 

The way income is divided between transfer earnings and economic rent depends on the characteristics of the particular factor’s supply. It is the elasticity or responsiveness of quantity supplied to a change in price which determines how much, if any, economic rent the owner of the factor receives as income in addition to transfer earnings. Here is a graphical illustration.

We have now seen that the pricing of factors of production in a market system simultaneously performs two important functions: one is to ration out scarce productive inputs so that they are used in their most effective application; the other is to determine the incomes of individuals and households who supply the services of these productive factors. The analysis to this point has assumed that factor markets are competitive and that the owners of factors consequently receive the marginal revenue products their inputs create. In real world market systems there are, of course, many imperfections in factor markets, just as there are in product markets. Consequently, factor prices and the incomes of factor owners may diverge from the ideal. How great this divergence is cannot be determined empirically with any precision. Some economists contend that factor markets do function quite smoothly and that factor prices adjust promptly to yield the kind of equilibrium adjustments of quantities supplied and demanded predicted by the theory just outlined. Others believe that factor markets, especially the labour market, are so affected by institutional and behavioural imperfections that they fail to achieve the equilibrium resolutions of demand and supply forces implied by the conventional marginal productivity theory.

The policy implications of this difference are important. If factor markets fail to assign the "right" prices to labour, land, and capital, there is no assurance that these factors are being employed in their most productive alternative uses. A case may then exist for government to intervene in the allocation of productive factors so as to improve the outcome, for example, by paying subsidies to encourage firms to employ certain types of workers or using tax incentives to steer individuals into particular kinds of employment.

Another implication of factor market failure is that the incomes received by factor owners, whether from supplying labour, land, or savings can no longer be justified on the grounds that they necessarily reflect the value of the marginal product their factor services produce. If this is so, then there may be grounds for arguing that the incomes received by individuals and households should be reallocated on the basis of some other criterion of what is correct or just. 

The extraordinary profit out of which rent arises is analogous to the extraordinary remuneration which an artizan (sic.) of more than common dexterity obtains beyond the wages given to a workman of ordinary skill. In the one case the monopoly is bounded by the existence of inferior soils, in the other of inferior degrees of dexterity.

—Samuel Bailey (1825)

The preceding analysis reflects the basic neo-classical theory of factor-pricing worked out in the late 19th century. Usually referred to as "marginal productivity theory", it suggests that in competitive factor markets, each factor of production will earn a return equal to its marginal revenue product (the market value of the last or marginal unit of output). That is, the wage rate will equal the marginal revenue product of labour, the return to land (rent) will equal the marginal revenue product of land, and the rate of interest will equal the marginal revenue product of capital. The usual processes of competition will cause the price of any factor of production to settle at the point where it is just equal to that factor’s productivity. Given sufficient time (i.e., in the long-run), each factor will earn a return proportional to its productivity. This is because producers will seek the least expensive way of producing their output. If they can obtain higher profits by using more machinery (capital) and less labour, they will substitute capital for labour. But using more capital causes the marginal revenue product of capital to fall, while using less labour causes the marginal revenue product of labour to rise. Each firm finds its best (most profitable) combination of factor inputs when the amount it spends on the last unit of capital brings in as much revenue as the last dollar it spends on labour or land. 

Once the terminology is understood and the basic axioms accepted none of this involves anything more than the most elementary logic chopping. But can the underlying assumptions be accepted? The neo-classical writers believed that they could. 

Writing at the turn of the century, J.B. Clark, an American economist, defended the claim that competitive factor markets resulted in a fair and equitable distribution of income in society. In his book, The Distribution of Wealth (1899), he began with the proposition that a fundamental principle of natural law was that each person is entitled to receive whatever that person creates. He went on to argue that if the wage rate, rent and interest are set at the value of the marginal product of each factor, and if all factors in each category are the same (all labour is the same, all land is the same, etc.) then any particular unit of a factor can be considered the marginal unit. If factors are not homogeneous within the usual categories, the differences among them will be reflected in earnings differentials which precisely represent the differences in their productivities. Following this line of reasoning affirms that since workers, land owners and capitalists are receiving the value of what they produce, the distribution of income in society must be in accord with the principles of natural justice.

This conclusion leads to a very conservative position with respect to government or other intervention in factor markets. It was, of course, rejected by socialist critics working in an intellectual tradition that denied Clark’s underlying assumption that private property is just. Do individuals have an unqualified claim on the productive ability they possess? It is easy to challenge this in the case of land. By what right does an individual lay claim to a piece of nature? To capital? Even to the ability to supply productive labour? These factor inputs are not produced by anyone, all are the products of nature. Does a highly trained surgeon have a claim on all the value he or she produces? Is not much of this productive ability a consequence of being (a) alive and (b) a member of society? How productive would anyone be if isolated from the rest of society?

It is still imagined by many that inequalities of income coincide broadly with inequalities of merit.

—Hugh Dalton (1925)
To Top

Income Distribution and Redistribution

How income is distributed in a society may be studied in two different ways. One is to look at how total income is divided on the basis of the various factors of production—labour, land, capital, and entrepreneurship. This functional distribution of income would be measured by summing all wages and salaries, rents, interest, and profits. Another approach would be to ignore how income is earned and focus instead on the size distribution of income among persons or households. 

The functional distribution of income may be derived from the national income data compiled by the national statistical agencies of most developed countries. Unfortunately the classification system employed by national income accountants does not mesh very well with the conventional concepts of wages and salaries, rents, interest, and profits used in economic theory. This makes it difficult to determine with any precision how the national income is divided among major groups in society, such as those who derive their income primarily from work and those whose income comes largely from the ownership of property. But in general it appears that in most developed countries something like 75 per cent of income is earned from employment, possibly more if a larger share of total earnings from farming and other unincorporated business is treated as labour income. This leaves less than 25 per cent as property income, much of which accrues to people who have some property, but who may also derive income from employment. Relatively few individuals depend mainly on income from investments (whether in stocks and bonds, real estate, or land) although in most of the highly developed economies the growing number of retired persons living on pensions and past savings is making this category more important than it used to be.

It is difficult to draw conclusions about historical trends in the division of income among such broadly-defined groups with any precision, but there is evidence of a modest shift of returns from capital to labour in most countries over the past several decades. This may be explained, however, by the continuing movement of labour out of agriculture and into manufacturing and services, a shift from employments where labour incomes have usually been relatively low into employments where labour incomes have tended to be higher. In most of the industrial countries, despite price supports and other policies aimed at supporting the agricultural sector, farm incomes have tended to decline relative to earnings in other employments. There has also been a tendency for the earnings of other types of unincorporated business to fall as a proportion of the total. Corporate earnings, on the other hand, have tended to rise, although corporate profits are highly variable in the short run due to their sensitivity to fluctuations in the general level of economic activity.

Information on the personal distribution of income has been collected by national statistical agencies on a fairly systematic basis since the end of World War II. Ideally, such surveys would provide a comprehensive view of the incomes individuals and households receive, such that all benefits, whether in the form of money or goods and services, would be recorded. This is not practical, however, and most official data are limited to money income received in the form of wages and salaries, net income from self-employment, income from investments, transfer payments from governments and miscellaneous items such as alimony and scholarships. "Free" services received from government agencies, goods produced within the household, capital gains and the kinds of job-related benefits enjoyed by many business executives and public employees are not included. This means that the published figures must be interpreted with caution, for it is not known how evenly or unevenly such unmeasured benefits are distributed throughout the population or how their distribution may be changing over time.

To Top

Measuring Income Distributions

Changes in patterns of income distribution are difficult to grasp if they have to be inferred from tables of detailed data. Because what is usually of interest in such comparisons is the degree of inequality in the distribution of income, it is much more convenient to use some kind of general statistic. While there are several others, the most frequently used measure of inequality in income distribution studies is the Gini coefficient named after the Italian statistician Corrado Gini, whose major work on the subject was published in 1912. A perfectly equal distribution of income in a population would have a Gini coefficient of zero. At the other extreme where all the income accrued to a single person the coefficient would have a value of 1. A graphical representation of the Gini is given by the Lorenz curve shown here.

Gini coefficients for the developed industrial countries range from low values of less than .3 for countries like Sweden and Norway to a high of over .4 for France, which has one of the most unequal distributions of income. Britain, Japan, Canada and the US have Gini’s in the .35 range.

Why are personal incomes as unequal as they are? There is no simple answer, but at least two plausible lines of explanation may be advanced. One is that wealth in the sense of tangible assets such as capital and land is very unequally distributed. Some families are rich because of income derived from property which is passed down from generation to generation. This is a relatively potent influence on the distribution of income in countries like the UK where large family fortunes have a much longer history than in newer countries like Canada or Australia. 

Although the unequal distribution of wealth is important in explaining the existence of the very rich, the inequalities separating most of the population into various income groups in all the developed countries arise mainly from differences in earnings from employment — and these differences must reflect the distribution of innate and acquired abilities already discussed. However, it is not known with any certainty how differences in income are to be attributed to the two different kinds of abilities. 

Whether income inequality is justified if it can be attributed to differences in productive capability, either innate or acquired is debatable. Advocates of income redistribution have argued that even if the distribution of innate abilities is unequal this does not mean that an unequal distribution of income is defensible. Before pursuing this issue of equity further, it may be noted as a matter of fact that all the advanced industrial countries have implemented policies and programs designed to reduce inequality in personal income distributions. Most of these measures have used the taxing and spending powers of government as instruments of redistribution policy. The personal income tax, with its progressive rate structure, is potentially a particularly powerful instrument for shifting the overall distribution of income in society toward less inequality. Many "welfare state" programs, such as old age pensions and mothers’ allowances, could also be expected to have this effect. Many of the benefits provided by these programs escape the reporting procedures used by national statistical agencies, and this means that conclusive evidence about the effects of government revenues and expenditures on the personal distribution of income cannot be produced. Such evidence as there is indicates that the personal income tax and the measurable transfer payments have reduced income inequality in most of the industrial democracies. A Canadian study shows, for example, that in the absence of taxes and government transfers the Gini for family income distribution in Canada in 1989 would have been .395, rather than .292. Yet there have been no dramatic changes in the overall pattern of income distribution in any of the developed countries during the past several decades. 

During the 1980s there were indications that in several countries, notably the US, Canada and the UK, the most notable trend in the distribution of income was an increasing polarisation, with the proportion of middle income earners shrinking, while the proportions at the extremes, the rich and the poor, were showing signs of increasing. In the 1990s this trend appeared to be weakening, but the possibility of a growing gap between the haves and the the have-nots has remained a matter of concern for those who believe inequality to be an important source of social unrest and instability.

It is better that some should be unhappy, than that none should be happy, which would be the case in a general state of equality.

—Samuel Johnson (1776)
To Top

The Goal of Redistribution Policy

Is inequality a problem? Should income be redistributed? These are complex questions that raise ethical and moral issues, some of which may lie outside the limits of economics as a discipline. Yet there are at least two elements in conventional economic theory which bear directly on this topic. One of these has already been implied in the treatment of factor pricing. There it was shown that in a well-behaved market system, the owners of productive factors will receive the value of the marginal product their resource inputs are capable of yielding. If that principle is combined with what has just been said about the sources of most personal incomes, it follows that to criticise the existing distribution of income in a market economy implies that there is something wrong with the factor pricing mechanism. Of course this may be the case. In the real world, unlike our simplified theoretical representation of it, there are many imperfections in markets, and certainly this is true in factor markets. Employers of factors may be expected to do everything possible to drive the prices they have to pay for factor inputs below the competitive equilibrium level and owners of such factors will be motivated to do everything in their power to push these prices above that level. If it could be shown that this contest is unequal, some corrective redistribution of income would be justified. Or would it?

Is there any reason to believe that the value of a worker’s marginal product should determine that worker’s level of income? If the worker received less, this could be seen as depriving the worker of something to which he or she is somehow naturally entitled. If the worker received more, obtaining income for which there was no corresponding output, this could interfere with the incentives that induce individuals to engage in productive effort, causing a reduction in total output and making everyone worse off.

Such considerations of economic efficiency are frequently invoked by participants in debates over income distribution, especially by those who oppose redistribution. An influential analysis of redistribution measures in the US over the period from 1950 to 1980, for example, claimed that welfare programs designed to reduce poverty and income differences had actually worked in the opposite direction by encouraging the poor to learn to live on welfare rather than on earned income. The implication of such a position is, of course, that individuals should be left largely to their own resources when it comes to earning a living.

But if some individuals are born with little in the way of innate abilities, or lack access to the schooling and other services needed to acquire learned abilities, how are they to manage? As we have seen, the Social Darwinists had no difficulty with this potential problem since the implication of their understanding of the competitive struggle was that such individuals should be allowed to perish on the grounds that this was nature’s way of ensuring that they could neither reproduce nor become a burden on the fit.

Today this extreme position is seldom openly expressed, although some of the more aggressive spokespersons for the far right occasionally come close to it. What is usually at issue now is the extent to which individuals should be insulated from market-based income determination. Since the 1970s there has been growing criticism of the existing welfare state provisions on the grounds that they have become too generous in their treatment of the economically disadvantaged or that they are ineffective. Arguments are also heard, however, in support of the status quo, or even for expanding social services into new areas such as day-care. Underlying all these differences is a fundamental intellectual problem. 

If income is to be redistributed, what should the ultimate objective of such redistribution be? For example, should society aim at eventually achieving an equal distribution of income, such that all individuals or families have the same level of income? It may come as a surprise to those who associate conventional economic theory with the politically conservative right to discover that there is an inherent case in favour of equality embodied in some traditional economic analysis. 

The marginal utility analysis discussed in Topic 3 suggested that the reason for the characteristic relationship between the quantity of a good demanded by a consumer and the price of that good could be explained by the phenomenon of diminishing marginal utility. This plausible proposition asserts that as an individual acquires successive units of a particular good, the additional benefit or satisfaction derived from additional (marginal) units of that good would decline. This principle has relevance to the subject of income distributions for, as the British economist, Hugh Dalton, put it, "in the language of common sense, the case against large inequalities of income is that the less urgent needs of the rich are satisfied while the more urgent needs of the poor are left unsatisfied....This is merely an application of the economist’s law of diminishing marginal utility." Transferring a dollar of income from a rich person (for whom the marginal utility of a dollar must be low) to a poor person (for whom the marginal utility of a dollar must be high) should certainly result in an increase in society’s total utility.

There are a number of difficulties associated with this analysis. One has to do with the great weakness of the utilitarian approach in general, the fact that utility is not measurable. Another is that individuals may have differing abilities to derive satisfaction from income. It is conceivable that the rich person in the above example can derive more satisfaction from a unit of income than the poor person because the rich person has more "refined" or more highly-developed tastes. Once differences in tastes are introduced, the utilitarian argument can be used to justify all kinds of income transfers in which marginal utilities of individuals are made equal and total satisfaction maximised, but with highly non-egalitarian consequences. On certain assumptions it can even be shown that total satisfaction could be maximised by transferring income from the poor to the rich! Utilitarianism consequently fails to provide a compelling guide to redistribution policy.

Other, more sophisticated theoretical models have been developed in the attempt to establish some rational basis for determining an ideal distribution of income but, like the older utilitarian analysis, all contain serious flaws which have prevented them from finding wide acceptance as a basis for policy. 

To Top

A "Just" Distribution of Income?

Modern economic analysis has generally sought to avoid involvement with ethical issues, focussing as much as possible on "positive" (scientific, non-judgmental) matters and leaving "normative" ("what should be") topics to others. But the lines separating economics from other disciplines are becoming increasingly blurred and this is particularly evident in areas involving the analysis of policy, including redistribution policy. In some respects this takes the discipline back to its roots, for it is relevant that Adam Smith, widely regarded as the founder of modern economics, occupied the chair of moral philosophy in the University of Edinburgh. In 1971 a present-day moral philosopher at Harvard University, John Rawls, published a remarkable book entitled A Theory of Justice. In it, Rawls sought to develop a theory of distributive justice devoid of the traditional utilitarian concepts. To achieve this, he invented a hypothetical reconstruction of how a society might have gone about determining a set of rules by which it would be ordered, a basic constitution which would be binding on all members of society for the rest of time. Rawls asks his reader to imagine an original situation in which the first members of society are trying to agree on the terms of such a constitution while being in a state of total ignorance about how the outcome would affect them as individuals. 

The principles of justice are chosen behind a veil of ignorance. This ensures that no one is advantaged or disadvantaged in the choice of principles by the outcome of natural chance of the contingency of social circumstances. Since all are similarly situated and no one is able to design principles to favor his particular condition, the principles of justice are the result of a fair agreement or bargain. (p. 12) More specifically: No one knows his place in society, his class position or social status; nor does he know his fortune in the distribution of natural assets and abilities, his intelligence and strength, and the like. Nor, again, does anyone know his conception of the good, the particulars of his rational plan of life, or even the special features of his psychology such as his aversion to risk or liability to optimism or pessimism. More than this, I assume that the parties do not know the particular circumstances of their own society. That is, they do not know its economic or political situation, or the level of civilization and culture it has been able to achieve. The persons in the original position have no information as to which generation they belong. (p.137) Given these conditions, Rawls argues that the rules of justice so agreed would provide that "each person is to have an equal right to the most extensive basic liberty compatible to similar liberty for others" and "social and economic inequalities are to be arranged so that they are both: (a) reasonably expected to be to everyone’s advantage, and (b) attached to offices and positions open to all." (p.150)

Under such a regime, whatever inequality existed would be viewed as fair and just by all concerned. It would be accepted that there is no injustice involved in some individuals having more than others, "provided that the situation of persons not so fortunate is thereby improved." In other words it is just that some have more if this has the effect of improving the position of those who are the least well off.

The intuitive idea is that since everyone’s well-being depends upon a scheme of cooperation without which no one could have a satisfactory life, the division of advantages should be such as to draw forth the willing cooperation of everyone taking part in it, including those less well situated. (p.15) Thus, individuals who get less than others in the distribution of income would not complain, presumably, because they would still be getting more than if they were living under a different set of rules. (More on Rawls)

The Rawls approach has been influential, particularly in the support it implies for redistributive measures designed to improve the lot of the worst-off members of society (assuming that such support does not incur the complaints of the better-off, an assumption which has become less and less plausible in recent times) But the Rawlsian theory has also been severely criticised, for example, by Robert Nozick, who attacked it from an extreme right-wing libertarian position.

Nozick’s basic position is that there is no way of determining what is a just distribution of income in any general sense. Instead, he would focus attention on the process by which incomes, and the subsequent distribution of incomes, is determined. In his view, justice has to do with how a particular distribution is arrived at. Consider the case of Bill Gates, founder of Microsoft, the enormously profitable computer software firm. Gates (at time of writing) enjoys a huge income, attributable to the millions of people who freely choose to buy his products. Is this "just"? Nozzick would say that it is, that the pattern of income distribution brought about through the free interaction of individuals all bent on improving their own lot in society must be considered just. If something happens to change the present distribution (perhaps Gates goes broke and loses everything because someone comes along with a better product) the new distribution is also just, assuming, of course, that it was brought about through just means. Thus, while the Rawls approach could lead to a judgement that a particular distribution was unjust and that, if the worst-off member of society could be benefited by transferring income from someone else, this injustice could be remedied, Nozick would object, arguing that depriving "someone else" of income, justly earned, was unjust and indefensible.

The Nozick approach is consistent with the view that a free market system is capable of generating outcomes which are not only efficient, but just. In this view, it is neither possible, nor desirable, to specify what rewards individuals deserve, any more than it is to dictate what goods and services should be produced. (More on Nozick.)

Having looked at the principles which underlie the functioning of particular markets, markets for goods and services and markets for factors of production, attention can now be turned to studying how the system as a whole works. If all markets function properly, it can be argued convincingly that the best thing to do is leave them alone. Interfering in their operation by imposing taxes, paying subsidies, enacting regulations, and redistributing incomes would probably be pernicious, causing the wrong things to be produced, the wrong quantities to be produced, discouraging individual effort, rewarding inefficiency. But what if markets do not work properly? Then the case for intervention may be made. 

There is no real necessity, and therefore no moral justification for extreme poverty side by side with great wealth. The inequalities of wealth though less than they are often represented to be, are a serious flaw in our economic organisation.

—Alfred Marshall

To Top