This is a simplified version of a paper to the Joint Conference between the Social Welfare Research Centre of the University of New South Wales and the New Zealand Planning Council, 10-11 November 1988, published in the Proceedings edited by Peter Saunders and Adam Jamrozik, as SWRC Report No 78, September 1989. This version was prepared in September 2006. Some remarks which were contemporary in 1988 have been omitted, as well as the more detailed analysis.
A short note on the alleged Equity-Efficiency Tradeoff
Keywords: Distributional Economics; History of Ideas, Methodology & Philosophy; Social Policy;
Introduction
The study of the distributional consequences of recent policy changes is a research program in its own right. Alas, the 1988 policy environment is antagonistic to serious research, perhaps more antagonistic than at any time since 1926 when the DSIR was founded. That means there is little useful research to use for such a paper as it risks the danger of being fragmentary rather than comprehensive, and open to misinterpretation.
In order to avoid the current fashion of opinion, servility, and superficiality as a substitute for serious independent analysis, this paper has set itself a limited task. Its main thrust is a review of the implicit objectives in much of the policy change and advocation, using the perspective of welfare economics. In particular it will become evident that much policy prescription which purports to be ‘value free’ is based upon values which may be self serving but are unlikely to reflect our traditional social objectives.
In addition the paper reports on some work which demonstrates that this issue is of considerable significance in the changing policy environment of recent years: It ends with a quick review of some research which indicates the limitations of the general framework which is used here, plus making the inevitable call for a sustained research effort.
The Objective of Efficiency
Here are some recent statements by a number .of New Zealand economists, all of whom have had some impact on policies and the implementation of policy in recent years. In each the policy objective which is referred to is ‘efficiency’, or some related production concept such as wealth maximisation. In each the objective is presented as though there is no question of its validity, that everyone would adopt the objective, and in this sense the implicit or explicit assumption of the objective is value free.
A system of incentives and sanctions is created through the market system to encourage efficient use of resources. It is generally argued that competition provides a set of incentives to promote good performance and a number of sanctions to penalise inadequate performance. The extent to which economic efficiency is maximised depends substantially on transaction costs. (Deane, 1988)
Economic efficiency should be the rationale underlying competition law. (Jennings and Begg, 1988)
Intervention in ‘market’ economies is inevitable because of the fundamental role of governments in establishing the institutional framework within which commercial activity is conducted. … These features of property rights underlie the system of incentives which operate in markets to ensure that resources are used more or less efficiently. (Jennings and Cameron, 1988)
It would perhaps reduce confusion if the term ‘competition policy’ were discarded in favour of ‘efficiency policy’ since economic efficiency should be the sole rationale for this type of intervention. (Kerr, 1988)
Proof of public benefit should not require applicants to prove the potential public benefit is distributed to any particular group. … The more emphasis that is placed upon the distribution of potential gains, as distinct from their realisation, the more political becomes the decision-making process. (Vautier, 1988)
Wealth optimising policy options. (Wilkinson, 1986)
It is harder to capture another feature of the articles from which the quotations are drawn. No other objectives are mentioned. It is not clear whether the writers think that the objective of efficiency is by itself sufficient, or whether they think in the particular circumstances they need only this objective.
Careful reading of these and other similar texts will indicate the concept of efficiency is not used rigorously, and is rarely defined. The concept is taken not only to be value free and important but almost primary without need for definition. This is particularly unfortunate, given that as we shall see the economic meaning is not the same as that in common usage.
We shall take it that the writers use the term with a common economic meaning, although interestingly that the third edition of The Macmillian Dictionary of Modern Economics (Pearce, 1986) does not define the term, emphasising its non-standard use. (It describes as Pareto optima what below is called Pareto efficiency.)
Greer (1988) writes
Economic efficiency is an archetypical teleological standard. By it we should incur total costs and gain total benefits in order to maximise net benefits. More precisely we can distinguish between allocation efficiency, production efficiency, and innovation efficiency.
Allocation efficiency is achieved when existing stocks of resources and technical knowledge are allocated to produce the collection of goods and services that buyers value most highly as indicated by their willingness to pay for them. …
Production efficiency requires that a given output be produced at the lowest possible cost in the light of known technologies and given resource prices. …
Innovation efficiency allows for shifts in demands (due to product innovations) and shifts in costs (due to production innovations) over time. (p. 7)
The use of the term which most closely corresponds to the common notion of efficiency is that of ‘production efficiency’ and its idea of using as little input as possible to get the given output However the aggregation of a variety of inputs (capital, energy, labour, materials, etc.) according to their costs involves an uncommon notion and, for instance, green critics of economics rarely accept that market prices give the correct weighting for aggregation.
This dispute amounts to the correctness of the notion of ‘allocation efficiency’. It is not our intention to review all of the value theory which explains why economists come to such a conclusion, and the caveats they place upon it
But Greer captures the issue relevant to this paper, when he writes that the output is assessed by that which ‘buyers value most highly’ as indicated by their willingness to pay. Thus the valuation of output is dependent upon the market demand of purchasers, and thereby their purchasing power. At which point we may ask where the income distribution, and related equity issues, fit in with the notion of efficiency. Is the notion value independent?
The Treasury on Efficiency
It is also difficult to demonstrate briefly the dominance of the notion of efficiency in the Treasury 1987 post-election briefing. It is true that other objectives are mentioned, but the concept ‘efficiency’ appears to be given more prominence and used more frequently than other objectives and criteria. At one point five pages are devoted to it, compared to two for equity. and just over one for the rest (pp 26-34).
Illustrating the degree of ambiguity in the use of the term ‘efficiency’, without definition, is discussed as an objective on page 26, with what amounts to a definition appearing on page 97.
The general conditions required for efficient production of goods and services can be stated relatively simply. Allocative efficiency is achieved when t_e prices paid for a firm’s output cover the cost of the resources involved, thereby drawing resources into production of goods in demand. Productive efficiency is achieved by producing the firm’s output at least cost. Finally, production should be expanded up to the point where the increase in revenue obtained from the additional sales just covers the additional costs incurred. (1987, p. 97) .
Many economists would dispute the Treasury definition, worrying about such ‘anomalies’ as unemployment and economies of scale, and noting that unlike Greer there is no recognition of dynamics. At issue for this paper though is that it is presented as value free, and the distributional assumptions implicit in the market demand are suppressed.
It is not necessary to pursue these points, but we note for later reference that the document talks of a ‘tradeoff’ between equity and efficiency (p. 33).
Utilitarian Welfare Economics
‘Efficiency’ is a word which has positive connotations. It would be strange to advocate inefficiency in a process unless one objects to the activity altogether. Yet as is evident from the discussion on Greer’s definition, the economist’s notion of efficiency is a special one, with a different meaning to and perhaps without all the positive connotations that the word has in normal usage. Moreover the way it is used in economics has implicit values underpinning it
This may seem strange to the non-economist, and it will be necessary to review part of the history of welfare economics to explain how economists arrived at this situation. The following illustration may be helpful.
Suppose we are at a point’ A’ and wish to reach ‘B’. Then we would want to get there as efficiently as possible. It might be more efficient, in some sense, to get from ‘A’ to point ‘C’. Do we take the more efficient option? Certainly not, for we want to get to ‘B’ not to ‘C’. Efficiency is about how to do something. By itself it does not tell us where we are going, or whether we want to go there. To put efficiency as a paramount objective, as the above quotations do, is to put the cart before the horse.
Traditionally neo-classical welfare economics centred around the notion that each individual’s behaviour could be characterised by a (mathematical) utility function in which the various commodities they consumed appears, and in which the more of each commodity the better (the higher the utility); The individual would select the commodities, paying for them subject to an income constraint, which would maximise the individual’s utility.
There are a number of objections to this characterisation of human behaviour, but these will be ignored here, although not in other contexts for such issues as social well-being, the environment, and the role of women. .
The utilitarian description of human behaviour had both a positive and a normative component The normative economics which underpins policy advice culminated in Arthur Pigou’s classic text The Economics of Welfare (1925). The approach involves interpersonal comparisons of individual utilities. In summary, and I simplify, individuals with a higher income are taken as having higher utility than individuals with lower incomes.
Experience has shown most people prefer to be rich than poor, and we can take their judgement about themselves as authoritative, at least for such economic analysis. However assuming that two persons on the same income may be treated as being at the same level of welfare is far less obvious. It may be said to be ‘not scientific’ in that it is not possible to envisage an experiment which would accept or reject the validity of interpersonal comparisons, except by slipping in further equally problematic assumptions. But, once this assumption of the validity of interpersonal comparisons is made, a formidable set of policy prescriptions could be adduced, including support for a progressive taxation and other redistributional strategies.
It was appropriate and understandable that the economists who followed Pigou should ask whether it was possible to derive policy prescriptions which were not dependent upon interpersonal comparisons. but without adding some other equally unsatisfactory non-scientific assumption.
The New Welfare Economics
The abandonment of interpersonal comparisons might have seemed doomed to failure but the following simple example shows otherwise. Suppose there are two people, one of whom has a bag of apples and the other has a bag of oranges.
Suppose they are permitted to trade, the result of which is that they each end up with a mix of apples and oranges, and. each considers themself now better off. The almost trivial policy conclusion is that permission to trade has lea to an improvement in each’s welfare, and therefore it is a better policy than would have been prohibiting trade, in this instance.
Note that it is not necessary to make any interpersonal comparisons of the welfare level of the two people to form a judgement. The policy conclusion is possible without this assumption, because in both cases welfare increased.
This situation was systematised in the notion of ‘Pareto efficiency’ (or Pareto optimality), named after the Italian economist Vilfredo Pareto. This referred to the circumstance where it was only possible to increase the welfare of one person by reducing the welfare of others. Conversely, if there was not Pareto efficiency the possibility existed of increasing the welfare of some people without detriment to the rest. All other things being equal it is better for an economy to be Pareto efficient, for if it is not then it should be possible to improve the welfare of some people, without detriment to others.
From this notion were developed a wide range of policy prescriptions of the form that market trading would, under.. certain circumstances, lead to a pareto efficient outcome. The required circumstances may not exist in reality. For the purposes of this paper let us assume that they do.
Even then Pareto efficiency has some difficulties. For instance it turns out that there is rarely a unique Pareto efficient.. point of exchange and production in the economy. That depends upon the initial allocation of resources; that is, upon; the income distribution. We shall have to come back to this issue, but there is an even more fundamental problem.
The fact of the matter is that the policy prescriptions based upon Pareto efficient criteria cannot answer some very pertinent policy questions. For instance they give no guidance upon whether a tax system should be progressive or not. This is not the same as the paradigm of new welfare economics giving a wrong answer. Rather there are policy issues central to economics that the paradigm could not, in this form, cover.
Every scientific paradigm has limits to its scope. For instance while economists may claim that they can forecast the rate of inflation, and we do and sometimes we get it wrong, economists make no claim to forecast the weather. This is not critical to the survival of economics, weather forecasting being outside the scope of any reasonable definition of economics. However it would be most peculiar if economists were to say that their theory did not allow them to predict inflation. .
Yet that is what happened with the new welfare economics. The income distribution, and policies to influence it, were. outside the range of issues it could comment on. It could say that for a given income distribution certain polices would lead to a Pareto efficient optimum, but that was not much better than meteorologists saying that given it were raining they could predict the ground would get wet.
One result of this was that Pigouvian welfare economics. continues to have an active role in economic and policy analysis. The work I have done on household equivalence scales and poverty is very much in that Pigouvian tradition, with the conscious and explicit use of interpersonal comparisons (Easton, 1976; 1980). We shall use another development shortly.
The Rise of Efficiency
The second possibility was that the inadequate ‘new’ or ordinalist welfare economies would add additional assumptions which would enable it to cover a wider range of relevant issues. As it happens the crucial assumption had as little scientific validity as that of interpersonal comparison, and perhaps less. Not surprisingly it slipped into the paradigm almost by accident.
What happened was that economists became interested in ‘compensation criteria’. (Hicks and Kaldor are the names associated with this development.) In simple terms. suppose there is a policy change which makes some people worse off, but those who are better pay (i.e. compensate) the losers. If the compensation takes place, and the winners are still better off after the side payments, then there has been a Pareto efficient improvement.
Now suppose that the change occurs, but the losers are not compensated. It could be said that such a change represents a ‘potential Pareto improvement’. For instance one could imagine a policy adviser telling a minister that the proposed . change was a potential Pareto improvement, but that consideration would have to be given to compensating the losers, or whatever if the minister had a different income redistributional objective in mind
It turns’ out that the potential improvement is. under certain assumptions, equivalent to an increase in real National Income (which for many purposes is much the same as increase in real GDP). The policy adviser could say more briefly that the change would increase National Income, but the minister would still have to consider compensation measures, if any. By doing so the adviser would still be working within the new welfare economics paradigm.
However it is but a step to advocate the policy because it increases National Income, and not to mention the compensation issue. This is equivalent to increasing national wealth, because wealth may be thought of as the discounted flow of income.
An alternative formulation is that the policy increases ‘efficiency’ because, assuming there is full employment, there is more income or output for the same inputs. Note that the expression ‘efficiency’ used here is no longer referring to Pareto efficiency. Indeed. even the formal notion of (non-Pareto) efficiency developed here has been replaced by a looser one, as a reading of the literature associated with the above quotations will show. Among the changes is that the writers may seem to refer to increases in the potential national income, rather than actual national income, since the increase in their ‘efficiency’ may be associated with additional unemployment. Sometimes they may be referring to the notion of production efficiency only. although the cost weightings may have little validity if there is unemployment.
What for our purposes is crucial, in all this muddle over the efficiency approach is that there is no requirement to examine the distributional consequences of change, and whether some individuals are made worse off. as a result of the implementation of the policies being advocated.
What is sometimes called ‘Hume’s Law’ is
that a dollar is a dollar. Hume’s law means that if two persons are bidding at an auction for a sea-side cottage and a poor homeless family is outbid by a wealthy family wishing to own a seaside weekender, the result of the bidding is efficient. The house has been placed in the hands of the most dollar votes. The effect of Hume’s law is to divorce consideration of the allocation of resources from consideration of the distribution of wealth [or income]. (Williams, 1988)
There is an irony that the name of Home should be associated with a principle which is so manifestly not a natural law. The doyen himself would perhaps have described it as a principle of conduct, and subjected its standing to a far more rigorous analysis than its users do today.
The Lorenz Analysis
By abandoning the compensation caveat, the policy advice has subtly introduced a judgement about income distributional fairness. To see what this assumption is we need to go back to the Pigouvian framework, albeit in a 1970s version.
At this point the original paper used Lorenz curves to explain how to compare distributions. In order to avoid the geometrical presentation, this version id going to skip the details and go to the nub of the issue as a preparation for the next section.
Take a particular income distribution. Suppose that some of the income of the rich is transferred to the poor (those with less income). We can unequivocally say that the latter distribution is less un equal than the former distribution. That involves no value judgements, other than common sense.
Similarly if there is a transfer of income from the poor to the rich there is an increase in inequality.
Note that if some income is transferred from those on in middle incomes to the richer, and some is transferred from the middle incomes to the poor, we cannot say there is an unequivocal increase or decrease in inequality without introducing a value judgement.
A more complicated situation occurs if the two income distributions also reflect different average incomes. Suppose that one with the higher average, also had lower incomes for the poor. How are we to compare the equity fo the two options.
John Rawls famously proposed a ‘veil of ignorance’, in which we do not know where we will be in the income distribution. So would we choose the one with the higher average income. If it was possible that we would be poorer. He argues we would not.
This implies we are totally risk adverse. Suppose but one person was worse off. Would we not gamble we were not that one person. Tony Atkinson has a different way of thinking about this – and a helpful one.
The Atkinson Analysis
Suppose we were to take a hundred dollars from one person and gave it to a second who was poorer. (The precise Atkinson measure is to someone who had exactly half income of the first). Using the previous analysis, that would be an unequivocal increase in equality.
But it might happen in this process of transferring some of the income was lost – through the deadweight loss of taxation. the disincentive effects, the costs of administration or whatever. The notion here is that a tax not only makes people worse off directly, but changes their behaviour. Consider someone who as a result of the higher income tax rate decides to work shorter hours, so there could be a loss of output to the economy as a whole. The measured loss would be even greater if she decided to cook and clean at home, now, rather than hire some on to do it. Much is made of this possibility, but it is also possible that the higher tax rate has her working longer hours in order to maintain her material standard of living – including the cleaner. (This possibility is played down by those who are advocating lower taxes on themselves.)
Suppose that we took the $100 from the rich and were able to give only $X to the poor where X was less than 100. Many people would judge whether this was a good or bad thing only if they knew how much X was.
For instance if X equalled 99 so that costs of transferring the money were 1 per cent, many people, perhaps most people, might well think that the transfer was justified and that the outcome was a better income distribution. On the other hand if X equalled 1, so that the costs of transferring were 99 per cent, most people might well think the change was unjustified. For different X there would be different proportions of the population who would come to different assessments.
This notion of the willingness to accept a loss for an income transfer can be generalised into an overall measure of the income distribution (called the Atkinson ‘eta’ coefficient). To simplify, at one extreme X is 100, and society is willing to tolerate no loss of output for a change in the income distribution. The other extreme X is 0, where equality of distribution is such a priority that the consequence may be a loss of total output. (It corresponds to the Rawlsian approach of evaluating the total income distribution from the position of the lowest income group.)
Practically most people’s preferences lie somewhere between. We might predict that on average X is higher (that is a greater reluctance to lose average income) in some societies, say the United States, than in others such as New Zealand. Leaving aside the stark stylisation of a society by otherwise identical individuals except for their income, this coefficient might be treated as a useful measure of attitudes to egalitarianism in a society.
It should be evident by now that the efficiency criteria on its own has a covert X of 100, a total reluctance to lose any output as a consequence of an income transfer). Thus, far from being distributionally neutral, advocation of efficiency as a policy objective involves an extremist distributional objective, which is only obscure because the advocates of efficiency fail to mention it.
The Equity-Efficiency Analysis.
( A short note on the alleged Equity-Efficiency Tradeoff)
This clarifies another puzzle. As mentioned earlier in the case of Treasury there are some who suggest there is a tradeoff between equity and efficiency. It appears to be a strange notion. Consider the announcement that the objective (of equity or whatever) was going to be pursued inefficiently. In normal parlance one pursues one’s objectives as efficiently as possible. But here the term ‘efficiency’ is not being used with a common meaning.
As far as one can judge, in this context the term efficiency can be replaced by national (or average) income or output, as follows from the above analysis. The essence of the alleged equity-efficiency trade off is that if there is any transfer there will be a loss of output (X will be less than 100) and average incomes will fall. That is true if the transfer is efficient’ in the nornmal meaning of the word.
In such circumstances, reference to an efficiency equity tradeoff is misleading. It is a sort of alliteration trick played by academics on undergraduates in exams to sort out the sheep from the goats.[6]
(However that fall may not always happen. Many would argue that over some ranges reducing inequality would increase average incomes, because of better human capital (from a better distribution of health, education, and like resources) and from improved performance from greater social solidarity (such as a higher degree of honesty and incorporation of externalities in individual behaviour).
In summary then, even a community which does not care about the degree of inequality may still practice some redistribution, because it may enhance average incomes.
It is also true that a community based upon the purest Rawlsian principles of evaluating only from the perspective of the poor, may allow income inequality if as a result of the inequlaity the income of the poor is higher.
There are at least two important caveats to this conclusion. First those who are rich may well have higher incomes under redistribution strategies. In such circumstances, without the veil of ignorance so they are aware of their income position, the rich may well advocate policies which reduce the national (average) income but are of benefit to them. This selfishness is not, of course, a peculiarity of the rich.
Second, the welfare analysis presented here is absolutest in material terms. If relativism was dominant, that is people cared only about the position in the income distribution, and not the income itself, the Utilitarian analysis would have to undergo substantial modification, as has been explored in recent years particularly by Amartya Sen.
The Trickle Down Effect
The ‘trickle down effect’ is a part of political rhetoric, but it deserves our consideration. The sort of policies which illustrate this phenomena might be that high taxation on the rich reduce incentives to work, invest, and save, which means less jobs and lower productivity for the poor. It may also mean less government revenue because of avoidance, and hence less social spending.
Thus despite the rhetoric, the trickle down effect may be important, and welfare enhancing. Presumably the objection to it is where the effect is small so that the poor are less well off relative to the rich. And sometimes it is used by opponents of redistribution in favour of the rich, as an ironic way of saying there will be no improvement in the lot of the poor.
Even so, I must report some bewilderment in an exchange in which I was tangentially involved. Accompanying a Listener article on the proposed December 1987 tax package which dramatically cut tax rates on high income recipients, was a cartoon by Trace Hodgson illustrating that in no uncertain terms the best the poor could expect was a trickle down from the tax cuts.I made no reference to this in the article (Easton, 1988).
In a far from considered response to the article, the Minister of Finance, wrote ‘I do not subscribe to the trickle down theory. That is to say, the idea that if you give massive benefits to the rich a bit will find its way down to the ordinary people’. (Douglas, 1988) Unfortunately there is no following sentence to tell us what the Minister does believe. Thus to what extent the clear government objective of ‘efficiency’, that is high real incomes, is moderated by equity is unclear.
Does Concentrating on Efficiency Matter?
All this would not matter if policies implemented on the basis of ‘national income efficiency’ were also Pareto. efficient, so that no-one was worse off as a result, and interpersonal comparisons were unnecessary. Unfortunately I know of no New Zealand research which examines the redistributive and output effects of changes on taxation, although there has been some work done overseas which could be applied here. However in at least two areas sufficient research has been done to indicate that often the income redistributive effects of a policy change are much larger than the gain in National Income.
One area is trade liberalisation. A 1979 review of the New Zealand literature found that while the models predicted, as might be expected given their assumptions, that wholesale trade liberalisation would increase national income these small changes, typically an increase of less than one per cent, were associated with dramatic changes in the profit to wage ratio, the regional pattern, and the size of individual sectors (Easton, 1980). . Further research has confirmed this finding (Philpott and Stroombergen, 1986).
These conclusions typically involve complex computational general equilibrium models. A similar conclusion, using a simpler more transparent model, is derived by Pickford (1986) when he examined the 1985 tariff cuts mainly on videos, motor vehicle parts and accessories, and microwave ovens, none of which were produced domestically. The effect of giving consumers a choice less distorted by erratic taxes was to increase national income by $11 million a year.
However the loss of tariff revenue to the general taxpayer was almost $72 million a year. That means the few purchasers of the tariff-cut items were $83 million a year better off and the general taxpayer $72 million a year worse off. Since the two groups do not coincide it follows that some people were worse off, although we cannot say who since we do not know how the revenue deficit was made up. The significance of the research for our purposes, is that it demonstrates the redistributive effects of the policy were far greater than the ‘efficiency’ effect.
The second research area where a similar conclusion is found comes from the general equilibrium modelling of labour market liberalisation (Easton, 1986). Suppose one occupational group in the community reduces its wages when some of its members are unemployed. That will generate additional demand for the workers and typically, more work and output generally. So there will bean increase in real National Income, and an increase in efficiency. However it is not evident as to whether the occupational group will share in this higher income. That will depend upon the increase in work relative to the cut in the pay rate, measured by the elasticity of demand for the occupation’s labour, the responses by other workers, the overall expansion in the economy, and the tax and social security arrangements.
This may appear to make generalisations possible, but the evidence points to the demand for most broad occupational groups being inelastic, and using this information in computational general equilibrium groups we find that while national income increases the income of the occupational group who takes the wage cut decreases.
For instance in March 1981 unemployment among professional people was 1.6 per cent. A 4.9 per cent drop in their salary rates would have eliminated this unemployment. Real national income would have risen .1 per cent, an increase in 1988 prices of $50 million a year. However the professionals would have had a drop in their real annual incomes equivalent to $132 million. Thus those who were not in the professional occupations would have been $182 million better off. Again the redistributional transfers are much larger than the output gain.
One piece of research on the microeconomics of liberalisation found effects which are not captured in the above studies. A study of five specific examples (Bollard and Easton, 1986) found little evidence for major gains in efficiency, measured by outputs relative to inputs. More important were a wider range of choice and better quality of service; It seems likely that these improvements are of most benefit to those on higher incomes, who have more discretionary spending. On the other hand liberalisation in the process of corporatisation, thus far not systematically studied, may indicate that the poor and those in peripheral regions got less choice and poorer quality as services were withdrawn.
There are hoped for ‘dynamic’ and entrepreneurial effects from liberalisation. If they do not appear, it may be the choice and quality effects will be ultimately the most important improvements from the policies. If so, they may well reinforce the distributive effects of the policies towards the rich.
The Politics of the Liberalisation
We are now beginning to see an insidious side to the advocation of the recent policies of liberalisation. They have been presented in terms of ‘value free’ efficiency, and of somehow being above politics. But the reality is that they have had an extremely value laden agenda. Policies are advocated providing they increase national income, irrespective of the distributional consequences.
Perhaps even that is not quite true. While the government has paid some attention to the needs of the poor, the main supporters and advocates of the policy thrust tend to support only those policies which promoted ‘efficiency’, and the interests of the rich. This is not difficult given that many of the policies of the past were probably willing to tradeoff national income maximisation for less inequality. But in some areas reform has been remarkably neglected.
Most evidently this applies to the financial sector, which has a number of advantages relative to the rest of the economy. These include the failure to institute a real capital gains tax, the continuation of taxation on the inflation premium of interest, the failure to impose GST upon financial services, the continuation of stamp duties which fall most heavily upon the small transactor, the failure to impose GST on foreign exchange purchases for consumption, and the financial market liberalisation without reform of commercial law. These have been of great advantage to the high fliers in the financial sector, but at the cost of the rest of the economy. It is surely no accident that the main advocates and supporters of recent reform have been in the financial sector, and who have notably failed to draw attention to these anomalies.
Indeed the situation where policies which promote ‘efficiency’ have a very small impact, while the redistributional gains are great, is likely to have the winners applauding the success of the policies out of proportion to the national gains – if any – because of the real benefits to them. In other words the economics of the liberalisation may be little more than the politics of rent seeking. Given the relative size of the redistribution of the national income gains, political analysts may find it more fruitful to describe recent events entirely in terms of rent seeking.
Complexity of People
Thus far we have treated all people as identical, except for having different income. However people have many differentiating characteristics, including sex, age, ethnicity, class, household situation and location. Suppose two societies had identical income distributions, but in males and females were evenly scattered through ti, while in the other all the males were at the bottom and the females at the top. Most would think the first society was fairer than the second (perhaps using a Rawlsian framework: under the veil of ignorance each would not know whether they were going to be a man or woman).
When I first wrote this paper there was some fragmentary evidence that the reforms were favouring the rich versus the poor. income distribution. Further research confirms this. (Easton 1996)(Easton 1996).
The paper also discussed unemployment and the income distribution. Again we now know more about what happened. (Easton 1996)(Easton 1996)
Future Directions
The paper finished discussing further research directions. Sadly these were not pursued. Whatever the economic logic against it, the pro-rich politics dominated the public discourse.
References
Atkinson, A. B. (1970) ‘On the Measurement of Economic Inequality’ , Journal of Economic Theory, Vol 2.
…. (1975) The Economics of Inequality, Oxford University Press, Oxford.
Bollard, A. E, and Buckle, R. A. (1988) Economic Liberalisation in New Zealand, AlIen and Unwin, Wellington.
Bollard, A. E. and Easton, B. H. (1986) Markets, Regulation, and Pricing, NZlER Research Paper No. 31, Wellington.
Deane, R. S. (1988) Privatisation and Beyond, address to the Annual Conference of the Electricity Supply Authorities, Wellington, 20 September.
Douglas, R. O. (1988) Response to Brian Easton’s Listener Article: ‘How Fair is the Flat Tax?’, Circulated Paper, 22 January. .
Easton, B. H. (1976) ‘Poverty in New Zealand: Estimates and Reflections’, Political Science, Vol 28, No. 2, December 1976.
…. (1980) ‘Three New Zealand Household Equivalence Scales’, New Zealand Statistician.
…. (1983) Income Distribution in New Zealand, NZlER Research Paper No. 28, Wellington.
…. (1987) Labour Flexibility, Wages, and Free Lunches, University of Melbourne, Department
of Economics Research Paper No. 180, Melbourne.
…. (1996) ‘Income Distribution’, A Study of Economic Reform: The Case of New Zealand, ed B. Silverstone, A. Bollard, and R. Lattimore, (North Holland Books: 1996) pp.101-138.
Greer, D. F. (1988) Efficiency And/Or/Versus/Equals Competition? paper presented to the AGM of the New Zealand Economists Association, September.
Jennings, S. and Begg, S. (1988) An Economic Review of Commerce Commission Decisions Under the Commerce Act 1986, paper presented for the Treasury, Jarden and Co. Ltd, Wellington.
Jennings, S. and Cameron, R. L, (1988) ‘State-Owned Enterprise Reform in New Zealand’, in Bollard and Buckle, op. cit., 121-152.
Kerr, R. L. (1988) ‘Time to Rethink Monopolies’, National Business Review, July, 8-9.
Pearce, D. W. (ed.) (1986) Macmillan Dictionary of Modern Economics, Macmillan, London, 3 ed.
Pigou, A. (1952) The Economics of Welfare, Macmillan and Co., 4th Edition.
Philpott, B. P. and Stroombergen, A. (1986) A General Equilibrium Analysis of Alternative Trade Regimes in New Zealand for 1990, RPEP Occasional Paper, Wellington.
Pickford, M. (1986) ‘Measuring the Effects of Tariff Cuts of 19 December 1985’, Massey Economic Papers, B8606.
Treasury (1987) Government Management, Post Election Briefing to the Incoming Government, Wellington.
Vautier, K. M. (1987) ‘Competition Policy and Competition Law in New Zealand’, in Buckle and Bollard (eds) op. cit., 46-66.
Wilkinson, B. (1986) Letter dated 31 January, to P. L. Read reported in Read The Treasury’s Fundamental Framework, Paper to the New Zealand Association of Economists, August 1986.
Williams, P. L. (1988) Why Regulate for Competition? Paper for N Z Centre for Independent Studies Conference, Regulating for Competition, Auckland, 7 March.