Welcome to TÁRKI homepage!

Microsimulation | Households, Work and Flexibility | Generational accounting | Gender Data Archive | TÁRKI—Andorka Library | TÁRKI Homepage

Tax Awareness…

 

Tax Awareness and Reform of the Welfare State

Results of a Hungarian Survey

by László Csontos, János Kornai and István György Tóth


The authors’ affiliations:

L. Csontos, Central European University, Budapest, died in 1997 at the age of 44, after the publication of the earlier, Hungarian version of this paper.

J. Kornai, Harvard University, Cambridge, Massachusetts and Collegium Budapest, Institute for Advanced Study, Budapest.

I. Gy. Tóth, TÁRKI (Social Research Informatics Center), Budapest.


1. Introduction

Reform of the welfare sector is on the agenda in Hungary, as it is in other post-socialist countries. Debate is taking place on the financing of the pension system, health care and higher education, on transformation of the social insurance system, and on the role of the welfare state. The actual fate of such reforms is always the result of a political process. Parliament has to pass new legislation to implement any important change. Before this takes place, there is a public debate between those who take various standpoints. This often prompts advocates of various mixes of public and private in the welfare system to cite opinion poll findings to support their ideas. However, the empirical basis on which they do so is not always clear. It is especially important under these circumstances to clarify what citizens know about these matters and what position they take on them. There is a great deal of literature on the transmission mechanisms that link together public opinion, political will expressed in votes, political parties and politicians, interest groups and their representatives, and the distortions in these mechanisms.

The survey we took among the population of Hungary in early 1996 was designed to answer two groups of questions. How clearly do Hungarians perceive the taxes that are levied on them in other words, to what extent are they tax aware? Is their picture of the link between tax payments and welfare services accurate or distorted? In addition, we wanted more information about the publics preferences for reform of the welfare system.

The research centred on a survey based on a questionnaire. The sample of 1000 was confined to the population of active age, and sufficiently representative of this. The interviewer spent about an hour in conversation with each respondent. One part of the interview was a customary survey by questionnaire. Answers were received about the characteristics of the respondents, and how informed and tax aware they were. The other part of the interview had elements of an experimental design. We wanted to know the magnitude of the tax burden respondents are ready to accept, in other words, how much income they are ready to relinquish to the state in return for various welfare services. This approach is analogous to the respondents entering a hypothetical market, where they have to say what price they are willing to pay for each public provision. The approach is closely akin to the contingent valuation method used to determine the demand for public goods. Respondents were invited to choose between alternatives, in an experimental decision-making situation, and from their choices we tried to deduce what their preferences would be.

We realize the survey has various limitations. With hindsight, we regret that the sample did not cover the inactive population as well. The hypothetical decision-making situations could have been described more clearly and fully (so that the reliability of the experiment's findings would be enhanced) if the time available for each interview had been substantially longer than an hour.

As mentioned already, the survey combined public-opinion polling techniques with contingent valuation techniques for revealing preferences. This method is not without its problems. As the interview progresses, the interviewer conveys to the respondent essential items of information that influence his or her opinion. However, this was precisely what interested us. What position do citizens take if, they know just what they forfeit in favour of the states welfare services, instead of just guessing rightly or wrongly?

The main purpose of our earlier papers was to publish the main facts from the data we had obtained. In this paper we go on to emphasise a few of the lessons to be drawn. It contains messages of two kinds. First we make factual statements, drawn from the survey. Of course, we cannot claim to have proved all of these statements by rigorous empirical tests. This was precluded by the limitations of the survey. However, we can claim that the credibility of these statements and the probability of them being true are increased because they rest on the survey's numerical results. Secondly, we put forward interpretations and opinions. These relate to the statements based on the survey. They are conjectural extensions of them, but interspersed with the authors suppositions, prior notions and value judgements. We have tried to word the study so that it remains clear where the research-based statements end and the authors' subjective opinions begin.

The issues addressed in the research are closely linked to the theory of optimal taxation and to the theory of preferences about taxation and state expenditures. These theoretical implications are not covered in this paper, which has the less ambitious purpose of providing a short account of the rich data of the first empirical survey. However, it also sets out to draw, some economic policy conclusions from the empirical data.

2. Weaknesses of tax awareness

The classical work that introduced the concept of fiscal illusions was written by Puviani (1903). This became known to economists through the work of James Buchanan, who also employed this theory for current phenomena. Cullis and Jones (1992) differentiated between two kinds of false awareness in citizens an optimistic. kind, which underestimates the tax load incurred from public services (and corresponds to fiscal illusions) and a pessimistic overestimates the tax burden, or perceives falsely and inaccurately the gains from the public service provided compared with the tax imposed.

Our findings suggest that both types of ignorance occur among Hungarian citizens. The survey supported the initial hypothesis that the majority of Hungarian citizens do not discern clearly the tax burdens placed upon them. They have trouble understanding the structure of the complex, intricate tax system. Many are not aware that various other tax-like contributions are levied on them apart from the charges openly described as taxes (personal income tax, for instance). These are earnings-related, contributions are partly paid by the employer and partly deducted from the employee's wages. When we asked what other types of deductions employers made from employees pay apart from personal income tax, 10.1 per cent of respondents said there were no other types, and 4.3 per cent replied that they did not know whether there were other types of deductions, or not. Of those who mentioned some kind of contribution, 16-24 per cent admitted that they did not know the rate. Interestingly, the knowledge about the employers contributions was more accurate.

The taxation levied on the general public includes indirect taxes on the price of goods consumed and services usedgeneral value-added tax (VAT), excise tax, and so on. Rather than analysing the overall data, we just asked about certain products as examples. Table 1 shows a breakdown of the replies for the tax content of gasoline and bread prices. Here and in later tables, we have differentiated between three grades of tax awareness. The correct reply is taken to be 100 per cent. If the response falls within a band of 75125 per cent, we have graded it largely correct. This interval is by no means very narrow. On the contrary, the measure is a lenient one. If the estimate is less than 75 per cent of the real figure, we have termed it significantly underestimated, and if it exceeds 125 per cent we have called it significantly overestimated. Underestimation fiscal illusions were strong in the case of gasoline. A high proportion of respondents significantly underestimated the tax content in the price of the product. With bread, the opposite was the case. Almost two-thirds of respondents greatly overestimated the tax levied on bread.

Citizens are not only misinformed about taxes. Most people have an uncertain knowledge of state expenditure and the true costs of certain ostensibly free public services. The observations are summed up in Table 2. Columns 1, 3 and 5 reflect the estimates and the actual figures for the total costs of a free or heavily subsidized provision. In these cases there are high proportions of respondents (58.8 per cent, 40.7 per cent and 62.6 per cent) who significantly underestimate the costs of the service, in other words, who live under strong fiscal illusions.

Columns 2, 4, 6 and 7 show the tax price of the various provisions per taxpayerthe annual average amount of tax imposed on the average taxpayer, in order to make the provision free or heavily subsidized. Underestimates are less frequent here, partly because the question came after respondents had guessed the total cost and learnt the right answer. With some tax prices, overestimation became more frequent and pronounced instead.

It is certainly surprising that the proportions of largely correct estimates of costs and their tax consequences are a mere 1325 per cent.

To sum up, the tax awareness of the general public seems to be weak. The knowledge of the vast majority is uncertain or wrong. Many people live under fiscal illusions, underestimating the tax burden required to maintain state services.

In our opinion, there is a combination of various factors behind this. One factor is that a false tax awareness arose under the socialist system. The tax burden on the population was concealed (notably the huge turnover tax built into the price of goods, and various levies on enterprises). People had the impression that the paternalist state was providing them with services free.

The tax reform of the reform-socialist period, which began with the introduction of personal income tax and VAT in 1988, made citizens at least partly aware of being taxpayers. However, the present tax system is highly complex and baffling to the average citizen. Politicians, government departments, the press, radio, television and schools have failed to explain adequately the connection between state provisions and taxation.

We do not claim that this situation is unique to Hungary or the post-socialist countries. The authors of the British survey say that failing to specify the tax consequences of changes in the level of public expenditure can lead to significant over-estimates of the popularity of such increases. The study provides a number of interesting results about the weaknesses and typical distortions of tax awareness in British citizens. It would be worthwhile conducting some international research, based on common methodological standards, that would allow firm comparisons to be drawn between tax awareness in Hungary and in other countries.

3. The distribution of preferences for state welfare services: a preliminary analysis

The survey included questions of various kinds that were designed to discover public opinion about the alternatives for reforming the welfare sector. Let us start by concentrating on one type. We described to respondents three possible kinds of institutional arrangements for various services currently provided by the state. These options appear in Table 3, where the questionnaire is quoted verbatim. Respondents could choose between centralized state solutions, a mixed structures and a market solutions. Before considering the distribution of the responses, there are some remarks to make about the questions themselves.

  1. The alternatives offered are a tiny fraction of the set of possible alternatives. Thousands of combinations of state, quasi-state or corporatist, and private elements can be envisaged, and, indeed exist in various countries. Since the choice we offered was very narrow, it could be, for instance, that even more people would have chosen a mixed structure if they had been offered alternative several combinations of state and non-state institutions and systems.
  2. The wording of the short explanation given for a market solutions was somewhat deficient, and therefore too extreme. Experts who advocate such solutions usually take it as self-evident that there is not unlimited freedom on the market for such services. The operation of the decentralized, autonomous institutions that provide the services will be constrained by law and supervised by state authorities, which regulate the prices for some services, perhaps provide state guarantees, and so on. The market solutions might have been more attractive to respondents if greates emphasis had been placed on these kinds of intervention and control.
  3. A further problem may have arisen because the interviewer combined the features of a questionnaire-based survey and a decision-making experiment. In a pure questionnaire-based survey, the interviewer does not convey any information to respondents. The aim is to find out how the respondents will react to the questions in the light of the information they already possess. Here the interviewer conveyed information to each respondent, saying what the tax consequences of particular state programmes would be, correcting, if necessary, the respondents mistaken information, and explaining the alternative mechanisms. However, the explanation attached to this last group of questions was not given in detail. The couple of minutes available proved very short for presenting complex alternative systems. Furthermore, these are not generally known alternatives. Some are merely conjectural possibilities in Hungary, taking forms not directly known to the Hungarian public, about which respondents had little personal experience. To have set up an accurate experimental situation, would have required providing more information, giving the respondent a clearer understanding of the hypothetical decision.

While pointing to the surveys shortcomings, we emphatically believe that the responses can be interpreted and evaluated. The numerical results obtained provide a picture of the main proportions, even though the accuracy of them can be questioned.

Table 4 gives a summary of the results. Support for the three types of provision differs from sphere to sphere. The state solution attracts the most support in the field of higher education. With hospital care, only a third of respondents support a centralized state solution while with the pension system, the proportion is just over a fifth. The mixed structures attract the most support in all three spheres, and are chosen by an absolute majority for the pension system. The level of support for the pure market solutions is 1218 per cent.

A fairly small proportion of respondents chose a pure state or a pure market solution for financing more than one of the three spheres. Those who consistently supported the centralized state solution, the status quo, in all three fields made up 9.8 per cent of the sample. The proportion consistently supporting a purely market solution for all three spheres was an insignificant 0.9 per cent. The vast majority preferred the mixed forms, combining state and market elements, for all three, for two, or for just one of the systems.

It can be concluded that there is broad support for reforms of the welfare sector that move away from the existing centralized, paternalist forms funded exclusively out of taxation, and introduce into spheres elements of decentralization, non-state institutions, and competition. However, the majority of the population shrink from extreme forms of laissez faire, whereby the state would withdraw from the welfare sector.

4. The dispersion of preferences and freedom of choice

We asked respondents how they thought HUF 100 of state expenditure are currently divided among four areas: (1) defence, police, administration and justice, (2) education and health, (3) other welfare services, and (4) economic purposes. We then asked what division they would prefer, and specifically, how the shares of education and health spending would fare. Only 6.8 per cent of respondents would leave the current proportion unchanged; 50.8 per cent would reduce it, and 42.4 per cent would increase it.

Let us look at Table 4 again. No majority view emerged about spending on universities or hospitals. Although the mixed structures for the pension system received majority support, respondents would certainly have differed if they had been asked to choose between various kinds of mixed structure.

There are some underlying lessons to draw here. The most important consideration is not the momentary percentage of public support for specific institutional solutions. There may always be measurement errors, and the views expressed may change. The most important aspect to recognize is the dispersion of the publics preferences, so that there is no alternative that attracts overwhelming support. That being the case, we do not consider the answer to impose the current majority view on those currently in a minority, who will then feel dissatisfied with the mechanism that emerges. The answer is to develop mechanisms that allow for as much choice as possible or widen the existing range of choices. Individuals should not be forced into joining specific schemes. This consideration alone provides strong grounds for replacing the state monopoly in the welfare system with competition among organizations of various forms of ownership, and parallel operation of various forms of finance, provision and insurance.

Public opinion is divided not only over the institutional forms the system should take, but over the desirable level of state funding. One scheme with which motorists at least are familiar is voluntary comprehensive insurance. In Hungary this normally includes a deductible part of each claim that the insurer does not paya co-payment by the insuredso that the loss is shared between them. The lower the co-payment, the higher the premium will be. We transferred this idea to the financing of hospital care and drug purchases. Let us take hospital care as an example. The social insurance system currently pays an average of HUF 1040 a month per insured (dependents as well as contributors) to cover the costs of hospital care. Let us assume that a private insurer undertook to finance hospital care in return for the same payment. Having explained this to respondents, we asked, Look at this table, where we have shown, alongside various monthly per capita insurance premiums, what percentage of the costs of hospital care you would have to pay. In this case, which would you choose? The distribution of the responses appears in Table 5. We followed a similar line of argument in the question about drugs (Table 6).

With the pension system, one of the questions we asked was whether the pension contribution should remain unchanged, be raised, or be reduced. The questions and the distribution of the responses appear in Table 7.

The lesson to be learnt from the quantitative choices in Tables 5, 6 and 7 is the same as for the qualitative choice between institutional forms. Peoples ideas are not uniform, but widely dispersed, so that the answer is not to make the procedures uniform. The public should not be forced to accept either a system of completely free provisions (and accompanying high taxes) or a system under which they cover all the costs themselves. Both qualitative (institutional) and quantitative alternatives should be offered where feasible, and people should be free to choose between them, according to their own systems of values and priorities.

5. The effect of tax and cost awareness on preferences

There are several factors that affect the preferences of individuals making choices about state welfare provisions. Among those whose effects our research allowed us to examine were (1) the respondents degree of tax and cost awareness, (2) the direct personal relevance to the respondent, and the respondent's (3) age, (4) education, (5) income, and (6) occupation.

We examined the effects of these factors separately, and we took some initial steps towards examining their combined effect, mainly through regression analysis. As a first step in the series of regression calculations, we sought to explain what factors induce people to adhere to or reject the status quo. This and the next two sections refer alternately to examinations of the effects of single factors and to the results of the regression analysis.

Starting with the effects of tax and cost awareness, the calculations suggest that the better a taxpaying citizen understands that there are no truly free gifts from the state, since taxpayers cover the cost of providing them, the greater the aversion shown towards the state solution.

The relation is rather stochastic, of course, as tax and cost awareness do not form the sole explanatory factor. Some citizens well aware of the link between tax and state provisions knowingly accept the tax burden those provisions entail. Some people are repelled by state paternalism because of factors other than tax awareness, while their knowledge about taxes is inaccurate. All we claim here is that there is a marked positive stochastic relation between tax awareness and antipathy to the state solution.

This shows up clearly in Table 8. The tax-awareness variable also proved to have strong explanatory value when the regression calculations for adherence to the status quo were made.

Although the interview lasted only an hour, it was in itself a substantive cognitive process for the respondents. This can be shown from a single example. As mentioned in Note 12, we only suggested two institutional alternatives for drug provision: maintenance of the present system of state subsidy, or abolition of the subsidy. Based on this knowledge, 75.1 per cent of respondents chose the state subsidy. Later, we presented respondents with a hypothetical graded system of premiums and co-payments (Table 6). This prompted them to consider that if they paid a higher premium (or by analogy, higher health-care tax or contributions), the deducted co-payment would be lower, while if the premium or tax were lower, the proportion of the co-payment would be higher. Once in possession of this extra information, the majority of respondents turned away from the principle of full state subsidy.

There was one more form in which we could study the relationship between tax awareness and the position taken by citizens. The following question was put in an earlier public-opinion survey: In the respondents opinion, did the state have an unconditional responsibility to (1) provide health care for the sick, and (2) provide a decent livelihood for the old? One of the authors of this study, János Kornai, made an objection to this formulation of the question, because an affirmative answer from the citizen was not associated with a conscious acceptance of the tax price of the ostensibly free state services. We think this procedure relies on a fundamentally false methodological premisethe unstated conviction that free lunches exist on a society-wide scale. If questionnaires compiled by this free-lunch methodology inquire about the state's unqualified obligations (for example, Does the state have an unconditional responsibility to care for the sick?), they conceal the elementary economic truths linking the costs of the state programmes with the tax paid by the respondents.

Table 9 compares the results of the earlier study with those of the new one. Although the question as formulated does not yield the true preferences in connection with state intervention, it was incorporated into the questionnaire for comparisons sake, in its original form. The following can be established from the table.

During the hours conversation, the interviewer gave some pieces of information about the actual tax costs of the ostensibly free programmes. Respondents were also made aware through the wording of the other questions that there is a link between the provision of a benefit by the state and the tax imposed on citizens. The interview obviously sufficed to alter the response distribution of the 1996 survey (Column 2) from the distribution found in the 1994 survey (Column 1) by 1016 percentage points. In other words, far fewer respondents in the second survey endorsed the vaguely expressed paternalist function of the state. More noteworthy still is Column 3, which presents the responses to a question radically different from the free-lunch methodology. Here we are not asking about a general responsibility of the state, but about a more tangible choice between state and not-purely-state institutional alternatives. There the proportion of support for the state solution is dramatically smaller, forming a minority of the respondents.

Assessment of the real support for state involvement is always a matter of strong controversy. There are a number of sociological surveys that show how respondents, when asked to give their opinion of the necessary involvement of the state, may take a strongly pro-state position. However, there are at least two reasons for scepticism about this. First, actual voting behaviour may differ strongly from declared opinions. Secondly, putting the question differently (for example, asking respondents to express their views on redistribution after revealing at least some tax prices) may lead to changed opinions. Although there are justified arguments for and against this assertion, we found support for it in our survey results. Pro-state enthusiasm seemed to cool as awareness increased.

There would clearly be further ways of examining the connection between tax awareness and these preferences. We could have asked respondents twice, under experimental conditions, which of the various institutional alternatives they would choose, offering, if possible, a wider choice. First they could answer on the basis of the information they possessed before the experiment, in other words, guided by their true or false tax awareness. Then the correct information could be given to them, and the question repeated. This would yield a direct information on the role played by the corrected tax information in revising the respondents initial decision. This is something we would like to do in future research.

6. The effect on preferences of direct personal interest or involvment

The hypothesis here is evident. When asked to make a choice on a question that concerns respondents personally they will be influenced in their choice by the impact the move will have on themselves and their families. There is a high chance of the alternative deemed personally beneficial being preferred. For brevitys sake, we will refer to this as self-interest. Various types and effects can be discerned.

What might induce respondents not to choose alternatives beneficial to them or their families? It might be a failure to discern what was personally beneficial, or it might be a readiness to override self-interest for altruistic reasons. These two reasons cannot be distinguished in our survey. It would require in-depth interviews to do so. However, as we questioned a large number of citizens, and it can hardly be assumed that all of them mistook their self-interest, some picture of the proportions between self-interest and altruism emerges. It was found that although direct self-interest affects the preferences on many, if not all questions, the effect is not very strong.

This is illustrated in Table 10 . Both columns support the hypothesis, especially the one showing the number of children currently supported. The more children a family contains, the greater the respondents desire that there should be free university or college education at the taxpayers expense. The calculus seems to be simple. The more children a family contains, the likelier it becomes that the saving obtained from free higher education will exceed the tax price of free education the respondent pays. Ultimately, such a family could receive a net donation from the other taxpayers, if each child benefited from higher education.

The table also shows that the correlation is not very strong. This may be because the respondents did not grasp the connection clearly although they were informed about the tax price of free higher education before they made the choice. On the other hand a respondent with several children may be certain that none of them will go to university, so there is no direct self-interest. However, altruistic considerations certainly played a part in many cases. An impressively high proportion of childless respondents, to whom the question is personally irrelevant at present, are prepared to pay the tax price of having free higher education. Even after subtracting the childless young, who may have children in the future, about a third of the respondents are prepared to give selfless support to free higher education. (We return to the question of altruism in Section 9.)

With reform of the pension system, the degree of personal relevance depends on the age of the respondents. Only 16 per cent of those under 24 believed in retaining the present pension system. This proportion rose by 34 percentage points per ten-year age group, reaching 24.4 per cent for those over 55. So there is a correlation, but it does not seem to be particularly strong. In the regression calculation on adherence to the pensions status quo, age again proved significant as an explanatory variable (with a significance level of 1.3 per cent). It is interesting that no relation appears between the choice of institutional alternatives for the pension system and cohabitation with parents or the parents age.

We conducted discriminant analysis for the characteristics of the group opposing institutional reform of hospital care (adhering to the status quo) and those supporting reform. One finding was that there are many more opponents of reform among those to whom the question is personally relevant, due to their own or their dependents illness. They presumably fear for the free nature of the service. Table 11 shows the results of another examination. The more disillusioning a personal hospital experience has been, the likelier a citizen becomes to want to replace the state solution with institutional alternatives that incorporate market elements.

The conclusion from the examination is that although direct self-interest influences peoples choices, it would be an indefensible simplification to explain the preferences solely by this, or more generally, by plain financial self-interest.

7. Other factors affecting preferences

The research supported the following two propositions:

The younger and more educated a respondent, and the higher his or her income, the likelier it becomes that he or she will support a shift away from the status quo, and choose one of the mechanisms that contain market elements, in other words that entail reform of the welfare sector. Adherents of reform are more frequent among entrepreneurs and professionals than among wage and salary-earners.

These propositions are confirmed by the regression analyses already mentioned. Almost without exception, the explanatory variables representing these factors proved strongly significant. They were incorporated into the estimation of the parameters of the regression equations, with the appropriate sign following from the above statements. The propositions are also supported by the discriminant analyses, which examined the dissimilar features of the groups adhering to and rejecting the status quo.

There are important economic-policy and political conclusions to be drawn from the propositions. They make clear which strata in society will accept ideas for reforming the welfare system more easily, and which will put up more resistance.

It is worth reflecting on what other factors affect those who adhere to the status quo, apart from degree of tax awareness, direct self-interest, age, standard of education, income and occupation. We are reduced to conjecture here. One obvious approach is a psychological one. The group wishing to retain the status quo should show a higher incidence of those who fear experimentation and innovation and do not want to risk change even if they do not like the existing conditions. Another factor may be the mentality of the many people who expect the state, solely or mainly, to look after their security. Value preferences may apply. The collective interest may be preferred over individual sovereignty, with a consequent aversion for market mechanisms, which are based on individual decisions. Past experience and allegiances may play a part. Those who found many features of the socialist system more attractive or advantageous than the present post-socialist transition to capitalism may cling, for intellectual, ideological and emotional reasons, to the old welfare system, as one of the main elements in the ancien régime.

Further research will be needed to test these and other possible conjectures.

8. Acceptance of the principle of self-reliance

One process complementary to curbing the paternalism of the state is to end the childlike treatment to which citizens have been subjected and restore their sovereignty. Citizens must assume greater independence, and concurrently, bear greater responsibility for themselves and their families.

The findings show that support and acceptance for the principle of self-reliance is dubious. Although some institutional arrangements, such as participation in voluntary private pension schemes, are expanding fast among certain segments of Hungarian society, opinion polls (sometime on the same subjects) show that this principle has not yet gained sufficient ground among Hungarian citizens a whole.

Asked how they would prepare for their retirement years, 50.7 per cent of respondents said they had not thought about it. This is an astonishingly high proportion. The disregard is not uniform, of course, but related to age, as Table 12 shows. To some extent it is understandable that those being further away from old age should think about it less, but sooner or later citizens will have to learn to prepare for their old age over several decades, throughout their earning lives. Only in part can this task be left to the state. A significant role must be played by various forms of private savings.

A sizable proportion of the population (although still a minority) already realize the need for self-reliance. This is expressed, for instance, when people take out various forms of voluntary insurance policies. When asked how they were preparing for their old age, 23.3 per cent of respondents mentioned that they are planning to take out an annuity or pension insurance policy, or join one of the voluntary pension funds.

With the reform of health-care financing, various types of misunderstandings of the principle of self-reliance arise. Many respondents thought there were only two alternatives: to have the state act as a universal insurer, or for individuals to cover the costs out of their own pocket. In fact developed, mature market economies have non-state insurance institutions, offering medical insurance policies, and covering most of the costs in exchange for a premium. The insured can choose between a lower co-payment with and a higher premium, or a higher co-payment with a lower premium. When this hypothetical option was explained to respondents, many of them were inclined to take advantage of it. (See Table 6.)

Those who had become acquainted with the idea of private insurance were more inclined towards reforms that would give greater scope for decentralized, non-state forms of insurance. Table 13 clearly links the number of private insurance policies that respondents hold with an increasing inclination to detach themselves from the insurance monopoly of the state.

9. Support for the principle of solidarity

We agree that the principle of solidarity should apply to welfare policy. On the one hand, this embodies the objective that society should care for those in need of such support. On the other, it requires redistributive taxation, whereby heavier taxes are imposed on those better able to bear them, provided this does not damage the spirit of enterprise, constrict innovation, reduce investment and savings, or dampen economic performance.

There is a view, usually advanced rather timidly, that most people are unwilling to make sacrifices for purposes of social solidarity. The argument goes that if Realpolitik sets out to apply this principle notwithstanding, it can only act as follows. While it imposes on the public a levy that it then spends on purposes of social solidarity, it draws peoples attention as little as possible to the fact that it is taking money out of their pockets to support others. This is the ostensible advantage of having a baffling system of contributions and redistribution that works without the publics conscious consent. If the public were asked for their permission, they might not have given it to that extent.

By the same argument, it is an advantage (not a drawback) of the so-called social insurance system in Hungary and many other countries that it inextricably mixes a true insurance system with elements of a redistributive system of tax-based financing. What happens is that different recipients of the same service (for instance, the same kind of health care) pay different prices (contributions) for it, depending on their income. This would be inconceivable for other products and services, including true insurance. Scrutiny of the economic content of the social-insurance contributions reveals that they irretrievably combine the functions of an insurance premium and progressive, income-related taxation.

This question raises some fundamental problems of political philosophy and ethics. Is it admissible to remove money from peoples pockets by subterfuge, even for noble purposes? Do the principles of the democratic political process leave room for the state, and the parliamentary majority governing it, to conceal from its voting, taxpaying citizens how it levies taxation and what it uses it for? Should it seek to apply the solidarity principle in this Machiavellian fashion, if it fears that the majority of citizens will not endorse its planned use of tax revenues? Or is it legitimate to confuse matters intentionally in the opposite case, when the majority of citizens, after receiving detailed information, can be said to be likely to approve of taxation for solidarity purposes anyway, so that it becomes superfluous to inform them and ask their permission? The authors of this study would answer all these questions with a qualified negative.

Whether the readers answer in negative, like the authors, or affirmative, they will doubtless agree that it is worth examining empirically whether such solidarity taxes need concealing at all. If the question were put to the public openly, would people refuse to pay?

Our research does not yield a clear picture of the publics point of view. For one thing, we were unable to unravel before them the insurance and redistributive strands in the present system, and present the problem comprehensively, within the space of an hours interview. Nonetheless, the findings have shed some light on this obscure area of public opinion and feeling.

As mentioned in Section 6, over a third of the respondents were prepared to accept the tax price paid for having free higher education, even if they had no children and could not expect any. Similarly, many people were willing to pay the redistributive element in the health-insurance system, even if they were not directly concerned at the time, and could feel more or less that they would be helping others, rather than themselves, by doing so.

We also tried to probe into peoples positions in a more direct way. Tables 14, 15 and 16 examine the willingness of respondents to pay taxes in cases where these expressly serve the purposes of social solidarity. In the first part of the interview we checked respondents tax awareness, and then informed them that a taxpayer pays an average of HUF 1100 a month for the purpose of social solidarity. His or her income could rise by this amount if this assistance through the state were to cease. We then presented the respondent with alternatives. Should the tax levied for this rise, stay the same, fall, or cease altogether?

Over half the respondents replied that the taxes paid for solidarity purposes should remain the same or increase. The groups that would reduce or abolish these taxes form a decided minority.

It is especially revealing to compare these three tables. With Table 14, the support for the alternatives is not affected by the degree of prior tax awareness. The tax is accepted or not in the knowledge that the present burden is HUF 1100 a month.

Table 15 shows that the proportion wishing to retain or increase the solidarity-tax level declines slightly as a function of higher level of education. This, however, swells the ranks of those who would distribute state revenues differently, not those wishing to reduce or abolish it.

It is especially interesting that Table 16 shows no correlation between readiness to show social solidarity and level of income.

From this can be drawn reassuring conclusions that reinforce the position expressed earlier. Even if those submitting welfare reforms admitted openly that they were calling for sacrifices for purposes of social solidarity, it would not be hopeless to expect the majority of the public to support this principle, or at least to refrain from opposing it.

10. The need to keep citizens better informed

Finally, we would like to point to one of the main lessons of the research: there is a need to improve the level of information among citizens. The conclusions are based on a Hungarian survey, but we are convinced they apply to the other post-socialist countries as well.

As on the content of their body of knowledge, citizens require several kinds of information. To start with, people need to realize that the money the state dispenses is the taxpayers, not state money. The change of system has made it possible and necessary to change this mentality. One of the main yardsticks for measuring the change is the prevalence and depth of the understanding that what one citizen receives from the state is paid for by the others. This insight should be promoted throughout the society.

Another necessary addition is for citizens to have far more specific and reliable information about the relation between taxes and state services. The better they appreciate what burden some favoured state programme places on the taxpayer, the greater the sense of responsibility with which they can take part in the political process.

Citizens should be as well informed as possible about the alternative ways in which the welfare system might be changed. One road to understanding the alternatives is to be familiar with variety of foreign experiences. Part of the problem at present is that the status quo, whether good or bad, is known to citizens from their own experience, but there has been little, or in some fields no chance to try other institutional forms of welfare provision. The greater the extent to which new forms are legalized, introduced and spread, so that their practical advantages and drawbacks emerge, the more substance the work of informing citizens about institutional reforms will gain.

Promoting greater tax awareness depends heavily on the behaviour and policy of various government agencies. The process will be advanced if the government and its constituent political parties ensure that they adequately inform and persuade the public when preparing practical measures of reform. When reforming the welfare systems, it is most important to provide patient and effective explanations of the reasons, purposes, and possible effects of the envisaged measures. Though it may not appear to be in their immediate interest, opposition parties can also contribute to more rational debate over taxation and spending, if their arguments link spending priorities with their cost-benefit implications.

In a democracy, the policies of the government and the parliamentary majority need to recognize the political preferences of citizens. However, this does not imply a passive acceptance in their political and moral approach. The government has a responsibility to lead. This includes, if necessary, influencing and attempting to change citizens preferences. Should the attempt fail, and the government cannot convince citizens that the changes it plans are justified, the electorate has the right to vote it out.

The aftermath of several decades of false teachings and harmful indoctrination cannot be dispelled from one day to the next. All democratic forces that subscribe to constitutionalism, parliamentary democracy and a market economy have a duty to contribute to this society-wide educational undertaking. The ways of bringing about a better understanding of the links between taxes, benefits and the redistribution process include reforming the curricula of schools and universities and conducting debates in the press and on radio and television.

The whole fabric of general government and the economic constitution should be clarified and made more transparent to taxpayers. Citizens tax awareness and the prudence of their choices will increase most when the budgetary and fiscal systems become more transparent and financial discipline consolidates. Then people will come to know from their own experience the new institutions and the new forms of property they can choose to replace the centralized, state forms of old. If all these favourable changes take place, the findings of the next survey of this kind will be more reassuring.


Gender Data Archive | Microsimulation | Flexibility | Generational Accounting


© TÁRKI – Last modified: August 21., 2003