The Economics of Taxation
[This essay is excerpted from Chapter 2 of Economics and Ethics of Private Property, newly published by the Mises Institute.]
First, I want to explain the general economic effect of taxation. This represents a praxeological analysis of taxation and as such should not be expected to go much beyond what has already been said by other economists.
To say there is nothing new to be stated regarding the economic effects of taxation is not to say that what there is would not be news to many. In fact, after surveying several popular economics textbooks it would seem that what I have to say is news to most of today's economists and students of economics. Insofar as these texts deal with the economic effects of taxation at all, beyond a purely descriptive presentation of various tax-schemes and their historical development, they are almost completely silent on the question of what the general effects of taxation are. Moreover, what in their discussion of the problem of tax-incidence these texts then say about the economic effects of specific forms of taxation is invariably flawed.
However, this state of affairs merely reflects a process of intellectual degeneration. As early as 150 years ago almost everything that should be understood today about the economics of taxation had been correctly and convincingly stated by such a prominent figure in the history of economics as Jean Baptiste Say in his Treatise on Political Economy.
In contrast to today's textbook writers, who assign the discussion of taxation to arbitrary places within the overall architectonic of their books, from the beginning Say correctly locates the phenomenon under the general heading "Of the Consumption of Wealth."
He then unmistakenly identifies and explains taxation as an attack on and punishment of the acquisition and production of property, which necessarily leads to a reduction in the formation of wealth embodied in such property and to a lowering of the general standard of living.
It is a glaring absurdity to pretend, that taxation contributes to national wealth, by engrossing part of the national produce, and enriches the nation by consuming part of its wealth.
Taxation is the transfer of a portion of the national products from the hands of individuals to those of the government, for the purpose of meeting public consumption or expenditure. Whatever be the denomination it bears, whether tax, contribution, duty, excise, custom, aid, subsidy, grant, or free gift, it is virtually a burden imposed upon individuals, either in a separate or corporate character, by the ruling power for the time being, for the purpose of supplying the consumption it may think proper to make at their expense; in short, an impost, in the literal sense.
Since such fundamental insights seem to have been forgotten, or at least no longer appear obvious today, let me, as my first task, present anew a praxeological account and explanation for Say's central argument and its validity, and in so doing refute some popular "counterarguments" claiming to show that taxation need not obstruct the formation of property and wealth. In light of this general explanation, I will then demonstrate the fundamental logical fallacy in the standard textbook analysis of tax-incidence.
That taxation — foremost and above all — is and must be understood as a means for the destruction of property and wealth-formation follows from a simple logical analysis of the meaning of taxation.
|Experience cannot beat logic.|
Taxation is a coercive, non-contractual transfer of definite physical assets (nowadays mostly, but not exclusively money), and the value embodied in them, from a person or group of persons who first held these assets and who could have derived an income from further holding them, to another, who now possesses them and now derives an income from so doing. How did these assets come into the hands of their original owners?
Ruling out that this was the outcome of another previous act of taxation, and noting that only those assets can be taxed that have not yet been consumed or whose value has not yet been exhausted through acts of consumption (a tax-gatherer does not take away another man's garbage but rather his still valuable assets!), three and only three possibilities exist: They come into one's possession either by one's having perceived certain nature-given goods as scarce and having actively brought them into one's possession before anyone else had seen and done so; by having produced them by means of one's labor out of such previously appropriated goods; or through voluntary, contractual acquisition from a previous appropriator or producer.
Only through these types of activities is one capable of acquiring and increasing valuable — and hence taxable — assets. Acts of original appropriation turn something which no one had previously perceived as a possible source of income into an income-providing asset; acts of production are by their very nature aimed at the transformation of a less valuable asset into a more valuable one; and every contractual exchange concerns the change and redirection of specific assets from the hands of those who value their possession less to those who value them more.
From this it follows that any form of taxation implies a reduction of income a person can expect to receive from original appropriation, from production, or from contracting. Since these activities require the employment of scarce means — at least time and the use of one's body — which could be used for consumption and/or leisure, the opportunity cost of performing them is raised. The marginal utility of appropriating, producing, and contracting is decreased, and the marginal utility of consumption and leisure increased. Accordingly, there will be a tendency to shift out of the former roles and into the latter ones.
Thus, by coercively transferring valuable, not yet consumed assets from their producers (in the wider sense of the term including appropriators and contractors) to people who have not produced them, taxation reduces producers' present income and their presently possible level of consumption. Moreover, it reduces the present incentive for future production of valuable assets and thereby also lowers future income and the future level of available consumption.
Taxation is not just a punishment of consumption without any effect on productive efforts; it is also an assault on production as the only means of providing for and possibly increasing future income and consumption expenditure. By lowering the present value associated with future-directed, value-productive efforts, taxation raises the effective rate of time preference, i.e., the rate of originary interest and, accordingly, leads to a shortening of the period of production and provision and so exerts an inexorable influence of pushing mankind into the direction of an existence of living from hand to mouth. Just increase taxation enough, and you will have mankind reduced to the level of barbaric animal beasts.
|"The marginal utility of appropriating, producing, and contracting is decreased, and the marginal utility of consumption and leisure increased."|
Straightforward as such reasoning may seem, there are a number of popular objections raised against it. For instance, from the side of economists who falsely conceive of economics as an empirical science that produces nothing but hypothetical explanations which invariably must be tested against empirical evidence in order to be validated (analogous to the situation in the natural sciences), the following argument is frequently heard: Empirically, it has been observed repeatedly that a rise in the level of taxation was actually accompanied by a rise (not a fall) in GNP or other measures of productive output; hence, the above reasoning, however plausible, must be considered empirically invalid. In fact, some empiricists of this sort go even further and make the stronger claim that taxation actually helps increase the standard of living as evidenced by the fact that some countries with once low standards of living and low levels of taxation now enjoy a much greater wealth with much higher taxes.
Both objections are simply confused. Experience cannot beat logic, and interpretations of observational evidence which are not in line with the laws of logical reasoning are no refutation of these but the sign of a muddled mind (or would one accept someone's observational report that he had seen a bird that was red and non-red all over at the same time as a refutation of the law of contradiction rather than the pronouncement of an idiot?).
As regards the stronger thesis, it is nothing but a beautiful illustration of the ever so attractive post hoc ergo propter hoc fallacy. From the fact that the correlated events of high taxation and wealth were to be observed later than those of low taxation and wealth it is inferred that increased taxation increases wealth. Yet to reason in this way is about as convincing as the argument — justly ridiculed by Say — that one can observe rich men consuming more than poor ones; therefore, their high level of consumption must be responsible for the fact that they are rich. Just as it follows from the meaning of consumption that this is impossible and that, on the contrary, the rich are not rich because of their high level of consumption but because they previously abstained from consumption and engaged instead in value-productive actions, so it follows from the meaning of taxation that mankind cannot have prospered because of higher levels of taxation but despite such a fact.
The weaker thesis — that experience would at least disprove any claim of a relationship between taxation and productive output that was negative by necessity — is also off the mark. The praxeological reasoning presented above does not at all rule out what empiricist economists falsely interpret as a refutation. In this earlier discussion the conclusion had been reached that the effect of taxation is a relative reduction in the production of valuable assets — a reduction, that is, as compared with the level of output that would have been produced had there been no taxation at all or had the level of taxation not been raised. Nothing was said or implied with respect to the absolute level of the output of valuable assets.
|Statistical studies here are entirely beside the point: they can neither help strengthen our analysis, nor can they ever be used to weaken it.|
As a matter of fact, absolute growth of GNP, for instance, is not only compatible with our earlier praxeological analysis, but can even be seen as a perfectly normal phenomenon to the extent that advances in productivity are possible and actually take place. If it has become possible through improvements in the technology of production to produce a higher output with an identical input (in terms of cost), or a physically identical output with a reduced input, then the coincidence of increased taxation and an increased output of valuable assets is anything but surprising. However, this does not in the least affect the validity of what has been stated about relative impoverishment resulting from taxation. With a given state of technological knowledge, though it may change over time, and taxation being what it is (a punishment of value-productive efforts), the level of productive output must be lower than the one that could have been attained with the same knowledge and no or lower taxation. Statistical studies here are entirely beside the point: they can neither help strengthen it, nor can they ever be used to weaken it.
Another theoretical objection which enjoys some popularity is that imposing or raising taxes leads to a reduction of income derived from the assets taxed; that this reduction raises the marginal utility of such assets as compared to what can be derived from other forms of activity; and thus, instead of lowering it, taxation actually helps increase the tendency to engage in production. For the usual case of taxing money assets this is to say that taxes reduce monetary income which raises the marginal utility of money, and this in turn increases the incentive to attain monetary returns.
This argument, to be sure, is perfectly true as far as it goes. However, it is a misconception to believe that it does anything to invalidate the relative impoverishment thesis that I have advanced. First of all, in order to keep the record straight it should be noted that even if it were true — as the just presented argument seems to suggest, albeit falsely as we will see — that increased taxation does not lead to a relatively lower output of valuable assets produced since it spurs a proportional increase in workaholism, it is still the case that the income of value-productive individuals has fallen. Even if they produce the same output as previously, they can only do so if they expend more labor now than before. Since any additional labor expenditure implies forgone leisure or consumption (leisure or consumption which they otherwise could have enjoyed along with the same output of valuable assets), their overall standard of living must be lower.
It now becomes apparent why the assumption that taxation can leave the productive output of valuable assets unaffected and exclusively cripple consumption is fatally flawed. If taxation reduces one's income (which includes that derived from present consumption and leisure), and given the universal fact of time preference, that is, that human actors invariably prefer present goods over future goods (that they cannot do without continuous consumption and can engage in lengthier, more roundabout methods of production only if a provision in the form of consumption goods has been made for the corresponding waiting period), then it necessarily follows that a person's effective rate of time preference must have been raised through this very act (the disutility of waiting must have increased), and that he will have to shorten the length of the structure of production as compared to the one that he otherwise would have chosen. Accordingly, his output of valuable assets available at future dates will have to be lower than would be the case otherwise.
If with lower or no taxation his income had been higher and his time preference schedule being given (whatever it happens to be at any particular point in time), he would have invested in lengthier production processes. As a consequence, his output of valuable future assets would have been relatively greater.
The error in the thesis that taxation can have a neutral effect on production lies in the fact that time preference is not taken into account. The argument presently under scrutiny is quite correct in pointing out that taxation implies a twofold signal: on the one hand the substitution effect working in favor of consumption and leisure and against work; and on the other hand the income effect of raising the marginal utility of the taxed asset.
However, it is false to interpret this simplistically as a mixed bag of contradictory signals — one in favor of and one against work — so that one can then state nothing of a categorical nature regarding the effects of taxation on production, and the question of whether or not taxation provides for a lower or a higher output of valuable assets must be conceived of as an entirely empirical one.
|"The error in the thesis that taxation can have a neutral effect on production lies in the fact that time preference is not taken into account."|
For in fact, the signal of taxation is not contradictory at all once it has been recognized that it is being sent to persons whose actions are invariably constrained by time preference. For such actors there exists not only the alternative between work and no work at all but also one between producing a valuable asset in more or less time-consuming ways. Invariably, they must also choose between obtaining an asset quickly and directly, with little waiting time involved, but at the price of having to resort to less efficient methods of production (the famous fisherman who decides to use his bare hands to catch fish in order to obtain it more quickly than by going through more roundabout methods of production), or obtaining it through more productive methods but then having to wait longer for them to bear fruit (the fisherman who, lured by higher future returns, decides to endure a longer waiting period and first builds a net).
However, given these choices, the message of taxation is completely unambiguous and unequivocal, and there can no longer be any question that the substitution effect must be thought of as systematically dominating any income effect: If there is not only the option of having something or not having it but also of having less of something sooner or more of it later, the double message sent through taxation is easily integrated and translated into one: reduce the waiting time; shorten the roundabout methods of production! By doing so, valuable assets will be obtained earlier — in line with their increased marginal utility.
Simultaneously, in shortening the waiting period, more room will be given for leisure — in line with its increased marginal utility. By reducing the length of roundabout methods of production the two seemingly contradictory signals stemming from taxation are simultaneously accounted for. Contrary to any claim of a systematically "neutral" effect of taxation on production, the consequence of any such shortening of roundabout methods of production is a lower output produced. The price that invariably must be paid for taxation, and for every increase in taxation, is a coercively lowered productivity that in turn reduces the standard of living in terms of valuable assets provided for future consumption. Every act of taxation necessarily exerts a push away from more highly capitalized, more productive production processes in the direction of a hand-to-mouth-existence.
It is not difficult to illustrate the validity of these conclusions if one considers the all-too-familiar case of taxing money assets. Such assets are only acquired and held because they can purchase other valuable assets at future dates. They have no intrinsic use-value at all (as in the case of a fiat paper money), or such use-value is insignificant compared to the exchange-value (as in the case of the gold standard where money also has an — albeit small — commodity value). Rather, the value attached to them is due to their future purchasing power. Yet if the value of money consists of representing other future available assets, the effects of taxing money becomes clear immediately.
Most importantly, along with increasing the marginal utility of leisure or consumption, such a tax increases the marginal utility of such future assets. This change in the constellation of incentives translates itself for an actor into increased attempts to obtain these assets more quickly, in less time-consuming production processes. The only production processes now that are systematically shorter than those of attaining future assets indirectly, via the earlier acquisition of money, are those of acquiring them through direct exchanges. Thus, taxation implies that barter trade will be substituted increasingly for the lengthier roundabout production method of monetary exchanges. But once again, resorting increasingly to barter is a regression to economic primitivism and barbarism.
It was precisely because production for bartering purposes yielded an extremely low output that mankind actually outgrew this developmental stage and instead increasingly resorted to and expanded a system of production-for-indirect-exchange purposes which, while requiring a longer waiting period, renders a far larger return of ever more and different assets drawn into the cash nexus. Every act of taxation means a coerced step backward in this process. It reduces output, decreases the extent of the division of labor, and leads to a reduction in social and economic integration (which, it may be noted, could never have become worldwide, were it not for the institution of indirect monetary exchanges).
|"Every act of taxation … reduces output, decreases the extent of the division of labor, and leads to a reduction in social and economic integration…."|
Furthermore, the general tendency towards increasingly adopting direct instead of indirect exchange mechanisms caused by every coercive seizure of money also has highly important consequences with regard to the methods of attaining money itself. Just as in the case of non-monetary assets, the increased marginal utility of money along with that of leisure-consumption also makes it relatively more attractive to acquire money in less time-consuming ways. Instead of acquiring it in return for value-productive efforts, i.e., within the framework of mutually beneficial exchanges, taxation raises the incentive to acquire it more quickly and directly, without having to go through such tediously roundabout methods as producing and contracting. On the one hand, this means that one will try more frequently to increase one's money assets by simply hiding them from the tax collector. On the other hand, a growing tendency will emerge to come into the possession of money through coercive seizure — either in the illegal form called stealing, or legally, by participating in the game called politics.
Having completed this general economic analysis of the effects of taxation, which today's economic textbook writers typically prefer not to deal with at all, let me now turn to what they typically do say about the effects of taxation under the heading of tax-incidence. In light of our previous analysis it will be easy to detect the fatal flaw in such accounts. Indeed, that one should fall headlong into error in dealing with specifics if one has not bothered to study the basics can hardly come as a complete surprise.
The standard account of the problem of tax-incidence most frequently exemplified by the case of an excise or sales tax goes like this: Suppose an excise or sales tax is imposed. Who must bear the burden of this? It is recognized — and I have of course no intention of disputing the validity of this — that in one sense there can be no question that consumers must take the brunt, and invariably do. For no matter what the specific consequences of such a tax are, it must either be the case that consumers will have to pay a higher price for the same goods and their standard of living will be impaired because of this, or it must be the case that the tax imposes higher costs on producers, and consumers will then be punished because of a lower output produced.
However, and it is with this that we will have to disagree sharply, it is then argued that whether or not the imposition of a tax harms consumers in the former or in the latter way is an open empirical question, the answer to which depends on the elasticity of demand for the taxed products. If the demand is sufficiently inelastic, then producers will shift the entire burden onto consumers in the form of higher prices. If it is highly elastic, then producers will have to absorb the tax in the form of higher costs of production, and if some section of the demand curve is inelastic and another elastic (this allegedly being empirically the most frequent case), then the burden somehow will have to be shared, with part of it being shifted onto consumers and another falling on producers.
What is wrong with this sort of argument? While it is couched in terms different from those used in my earlier analysis, one can hardly fail to notice that it merely restates, on a somewhat more specific level of discussion, what has already been demonstrated as false on a more general level: the thesis that taxes may or may not reduce productive output; that there is no necessary connection between taxes and productive output; and that it must be considered empirically possible that a tax may affect consumption exclusively while production remains untouched. To assume, as the textbook-account of tax-incidence does, that taxes can be shifted forward, either totally or partially, onto consumers is simply to say that a tax may not negatively affect production. For if it were possible to shift any amount of a tax forward onto consumers, that amount would represent a "non-production tax," a tax exclusively on consumption.
In order to refute the typical textbook analysis, one could simply go back to our previous discussion that resulted in the conclusion that any tax imposed on people constrained by time preference must negatively affect production above and beyond any negative consequences that it implies for consumption. However, I will choose a somewhat different route of argument here in order to make essentially the same point and thereby establish the more specific thesis that no amount of any tax can be shifted onto consumers. To assume otherwise is to assume something manifestly impossible.
|"A growing tendency will emerge to come into the possession of money through coercive seizure ¡X either in the illegal form called stealing, or legally, by participating in the game called politics."|
The absurdity of the tax-forward-shifting doctrine becomes clear as soon as one tries to apply it to the case of a single actor who continuously acts in both roles — that of a producer and a consumer. For such a producer-consumer, the doctrine amounts to this proposition: If he is faced with an increase in the costs of attaining some future good — an increase, that is, that he himself perceives as a cost-increasing event — then he shifts these higher costs onto himself in such a way that he responds by attaching a correspondingly higher value to the good to be obtained, thereby restoring his old profit-margin, thus leaving his role as producer unchanged and unimpaired, and requiring restrictive adjustments exclusively in his role as a consumer. Or, formulated even more drastically, insofar as his value-productive efforts are concerned, a tax does not make any difference for an individual, because he just starts liking the to-be-produced good correspondingly more.
Plain reasoning reveals that what generates such absurdity is a fundamental conceptual confusion: The forward-shifting doctrine arises from not recognizing that in one's analysis one must assume that demand is given — and that this must be assumed because it in fact is given at any point in time. Any analysis that loses track of this is flawed, for if one were to assume that demand had changed, then everything would be possible: production might increase, decrease, or remain unchanged. If I am a producer of tea and tea is taxed and if it is assumed that the demand schedule for tea rises concurrently, then, naturally, it is possible that people are now willing to pay a higher price for tea than previously.
Yet this is obviously not a forward shifting of the tax but the result of a change in demand. To present this possibility under the heading of tax-incidence analysis is plain nonsense: it is in fact an analysis of the entirely different question of how prices are affected by changes in demand and has nothing whatsoever to do with the effects of taxation. The confusion here is on as grand a scale as that which one would encounter if someone were to "refute" the statement that one apple and another make two by saying "No, I have just added another apple, and look, there are not two but three apples here." It is hard to get away with such nonsense in math; in economics a doctrine hardly less absurd is the orthodoxy.
Yet if one is logically committed to assuming demand to be given whenever one tries to answer the question whether or not a tax can be shifted forward, every tax must be interpreted as an event that exclusively affects the supply side: it reduces the supplies at the disposal of suppliers. Any other conclusion would amount to a denial of what had been assumed from the outset — that a tax had indeed been imposed and perceived as such by producers. To say that only the supply curve is shifted whenever a tax is extracted (while the demand curve remains the same as before) is to say nothing else than that the entire tax-burden must in fact be absorbed by the suppliers.
To be sure, the leftward shift of the supply curve would cause prices to rise and consumers would naturally be harmed by having to pay these higher prices and by only being able to afford a smaller amount of goods at such a price. Yet that consumers will invariably be hurt by taxes has of course never been doubted as one should recall. However, it is a misconception to think that this higher price is a shifting of the tax burden from producers to consumers. Rather, consumers are hurt here "only" by harm being done to producers who, despite higher prices charged for their supplies, must bear the brunt.
One must ask oneself why, if an entrepreneur could indeed shift any amount of the tax-burden away from himself and onto consumers, he would not have already done so by voluntarily imposing a tax on himself instead of waiting for the actual coercive tax to come along! The answer is clear: At all times he is constrained in his price-setting activity by the actual given demand. The price set by any entrepreneur is set with the expectation that a price higher than the one actually chosen would yield a lower total revenue. Otherwise, if he expected a higher price to bring about a larger revenue he would raise it. As long as an entrepreneur expects the demand to be inelastic within the region of any price-range under consideration, he will take advantage of this and choose the higher price. He stops raising prices and settles for a specific one because his expectations are reversed and he anticipates the demand curve above this price to be elastic. These expectations regarding inelastic and elastic portions of the demand curve are not at all changed if the entrepreneur is faced with a tax. Then as now he expects higher prices to produce revenue losses. Thus, it is obviously out of the question to argue that he could escape the burden of the tax. In fact, if as a consequence of the reduced supply the price now rises, this upward movement must be into an elastic portion of the demand curve, and the entrepreneur thus must assumedly pay the full price of it in the form of reduced revenue. Any other conclusion is logically flawed.
Only if the entrepreneur expects a change in demand occurring simultaneously with taxation could he change his price without thereby incurring losses. If he expects demand to have increased, for instance, such that there will now be an inelastic rather than an elastic stretch of the demand curve above the presently going price, he will be able to raise it without punishment. Again, this is not a forward shifting of the tax. This is increased demand. With or without the tax the entrepreneur would have acted in precisely the same way. The tax has nothing to do with such price changes. In any case, the tax must be paid exclusively and in full by the suppliers of the taxed goods.15
[Continued in "The Sociology of Taxation"]
Hans-Hermann Hoppe is professor of economics at the University of Nevada at Las Vegas. Send him mail. This article is excerpted from Chapter Two of Hoppe's Economics and Ethics of Private Property, newly published by the Mises Institute ($28). Comment on the blog.
 Exclusively descriptive analyses of taxation are given, for instance, by Paul Samuelson, Economics, 10th ed. (New York: McGraw Hill, 1976), chap. 9; Roger L. Miller, Economics Today, 6th ed. (New York: Harper and Row, 1988), chap. 6.
 Jean Baptiste Say, A Treatise on Political Economy (New York: Augustus M. Kelley, 1964), pp. 446 — 47.
 Ibid., p. 446; on Say's economic analysis of taxation see also Murray N. Rothbard, "The Myth of Neutral Taxation," Cato Journal (Fall, 1981), esp. pp. 551 — 54.
 See on this also Murray N. Rothbard, Man, Economy, and State (Los Angeles: Nash, 1970), chap. 12.8; idem, Power and Market (Kansas City: Sheed Andrews and McMeel, 1977), chap. 4, 1 — 3.
 See Say, A Treatise on Political Economy, p. 448.
 See on this point also Rothbard, Power and Market, pp. 95f.
 One might object here that the tax receipts will come into someone's hands — those of government officials or of governmental transfer-payment-recipients — and that their increased income, resulting in a lower effective time preference rate for them, may offset the increase in this rate on the taxpayers' side and hence leave the overall rate and the structure of production unchanged. Such reasoning, however, is categorically flawed: For one thing, insofar as government expenditure is concerned, it cannot be regarded as investment at all. Rather, it is consumption, and consumption alone. For, as Rothbard has explained, [i]n any sort of division-of-labor economy, capital goods are built, not for their own sake by the investor, but in order to use them to produce lower-order and eventually consumers' goods. In short, a characteristic of an investment expenditure is that the good in question is not being used to fulfill the needs of the investor, but of someone else — the consumer. Yet, when government confiscates resources from the private market economy, it is precisely defying the wishes of the consumers; when government invests in any good, it does so to serve the whims of government officials, not the desires of consumers. (Man, Economy, and State, pp. 816 — 17)
Thus, government expenditure, by definition, cannot be conceived of as lengthening the production structure and hence as counterbalancing the taxpayers' raised time preference rate. On the other hand, as for the transfer expenditures made by the government (including the salaries of bureaucrats and subsidies to privileged groups), it is true that some of this will be saved and invested. These investments, however, will not represent the voluntary desires of consumers, but rather investments in fields of production not desired by the producing consumers…. Once let the tax be eliminated, and … the new investments called forth by the demands of the specially privileged will turn out to be malinvestments. (Power and Market, p. 98)
Consequently, transfer expenditures also cannot be conceived of as compensating for the fact that taxpayers shorten the length of the production structure. All such expenditures can do is to lengthen the structure of mal-production. "At any rate" concludes Rothbard, the amount consumed by the government insures that the effect of income taxation will be to raise time-preference ratios and to reduce saving and investment. (Ibid., p. 98)
 See for such — irrelevant — empirical studies regarding the relative importance of income vs. substitution effects George F. Break, "The Incidence and Economic Effects of Taxation," in The Economics of Public Finance (Washington, D.C.: Brookings, 1974), pp. 180ff.; A.B. Atkinson and Joseph E. Stiglitz, Lectures on Public Economics (New York: McGraw Hill, 1980), pp. 48ff.; Stiglitz, Economics of the Public Sector (New York: Norton, 1986), p. 372.
 Here once again what has already been explained in a somewhat different connection in note 7 above becomes evident: why it is a fundamental mistake to think that taxation might have a "neutral" effect on production such that any "negative" effects on taxpayers may be compensated by corresponding "positive" effects on tax spenders. What is overlooked in this sort of reasoning is that the introduction of taxation not only implies favoring nonproducers at the expense of producers. It simultaneously changes, for producers and nonproducers alike, the cost attached to different methods of attaining an income, for it is then relatively less costly to attain an additional income through nonproductive means, i.e., not through actually producing more goods but by participating in the process of noncontractual acquisitions of already produced goods. If such a different incentive structure is applied to a given population, then the length of the production structure will necessarily be shortened, and a decrease in the output of goods produced must result. See on this also Hans-Hermann Hoppe, A Theory of Socialism and Capitalism (Boston: Kluwer Academic Publishers, 1989), chap. 4.
 See for instance William Baumol and Alan Blinder, Economics: Principles and Policy (New York: Harcourt Brace Jovanovich, 1979), pp. 636ff.; Daniel R. Fusfeld, Economics: Principles of Political Economy, 3rd ed. (Glenview, Ill.: Scott, Foresman, 1987), pp. 639ff.; Robert Ekelund and Robert Tollison, Microeconomics, 2nd ed. (Glenview, Ill.: Scott, Foresman, 1988), pp. 463ff. and 469f.; Stanley Fisher, Rudiger Dornbusch, and Richard Schmalensee, Microeconomics, 2nd ed. (New York: McGraw Hill, 1988), pp. 385f.
 On the impossibility of a pure consumption tax see also Rothbard, Power and Market, pp. 108ff.
 Baumol and Blinder, Economics: Principles and Policy, p. 636, present the demand curve as changing in response to a tax.
 To avoid any misunderstanding then: Insofar as the textbook analyses of tax-incidence point out this fact they are of course entirely correct. It is the interpretation of this phenomenon they give which is fundamentally confused!
 See on this point also Rothbard, Man, Economy, and State, p. 809.
- 15. Should a tax not immediately affect supply at all, as can happen in the short run, then it follows from the above analysis that the price charged will not change at all. For to raise it in response to the tax would once again imply pushing it into an elastic region of the demand curve. In the long run the supply will have to be relatively reduced and prices must move into this region. In any case, no forward shifting takes place. See on this also Rothbard, Man, Economy, and State, pp. 807ff.; idem, Power and Market, pp. 88ff.