Sraffa on taxable income and its implications for fiscal policy

E S Levrero

doi:10.1093/cje/bex050

Levrero, E S

2017-12-05 00:00:00

Abstract The aim of this paper is to reconstruct Sraffa’s analysis of taxable income and the subsistence wage as traceable in his unpublished manuscripts and then look at some of its implications for policy analysis—especially with regard to taxation and an operational setting of minimum wages and subsidies for the poorest sectors of the population. It also aims to point out how these kinds of implications are usually traceable (at least until recently) in the literature on taxation since, as Sraffa wrote in November 1927, ‘public finance (the only practical part of economics) acts on the theory of the surplus, by necessity’. I. Introduction In his early manuscripts Sraffa drew attention to the fiscal practice of exempting the subsistence wage from taxation and criticized Fisher, Marshall and Pigou for not subtracting what is needed to replace ‘the destroyed cells of the labourers’ (Einaudi, 1916, p. 205) from gross income when defining net income. This kind of criticism especially emerges in Sraffa’s manuscripts on Fisher’s definition of net income, according to which, unlike the interest rate, ‘net wages’ should be calculated by deducting the ‘disservices’ (that is, work-related pains and other discomforts) from the value of the flow of goods given to workers, but not the ‘expenses’ needed for the workers’ reproduction, since net national income should be conceived of as the mass of goods providing a utility, and workers receive satisfaction even from that part of wages that is indispensable for their reproduction, i.e. from the subsistence wage. According to Sraffa, such a definition by Fisher is contradictory since it does not consider that capitalists also receive benefit from replacement of their capital so that a coherent definition of net national income should include the ‘depreciation expenses’ for both labour and capital or should exclude those expenses in both cases. A calculation of the ‘depreciation expenses’ for labour requires, of course, a definition of the subsistence (or cost component of) wages on which Sraffa first reflected in the years 1922–26 mainly on the basis of the works of Cannan ([1917] 1967), Loria (1909), Marshall ([1920] 1949), Pantaleoni ([1889] 1911), Pigou (1921) and Ricca Salerno (1900). When, in the years 1927–29, Sraffa rediscovered the classical or surplus approach and succeeded in elaborating his ‘first equations’ (see Garegnani, 2005; Gerkhe and Kurz, 2006; Kurz, 2014), his reflections on the cost and surplus components of wages were enriched by a deeper analysis of Smith’s, Ricardo’s and Marx’s notions of costs and their definition of the ‘necessary’ wage as being one able to reproduce an efficient worker according to the demographic and social-historical conditions of the time. The aim of this paper is to reconstruct Sraffa’s analysis of taxable income and the subsistence wage as traceable in his unpublished manuscripts and then look at some of its implications for policy analysis—especially with regard to taxation and an operational setting of minimum wages and subsidies for the poorest sectors of the population. It also aims to point out how these kinds of implications are usually traceable (at least, until recently) in the literature on taxation since, as Sraffa wrote in November 1927, ‘public finance (the only practical part of economics) acts on the theory of the surplus, by necessity’ [D3/12/11: 56]. 2. The peculiarity of labour as a commodity A part of Sraffa’s 1922–26 unpublished manuscripts1 dealt with wages and the definition of net income. He painstakingly annotated here any reference to the notion of subsistence as discussed by Cannan, Marshall, Pantaleoni and Ricca Salerno, as well as any distinction between the subsistence and surplus wage. There are also short notes on the different wage theories which (following Cannan [1917] 1967) Sraffa classified as the subsistence theory, the theory of demand and supply (namely, the wage fund doctrine) and the ‘product theory’. These manuscripts seem to reflect an initial research project of Sraffa on wages and income distribution written along Marshallian lines. In a scheme of the work to be done, Sraffa refers to Marshall’s analysis of the peculiarity of labour as a commodity (see D1/58: 4)2, while in another manuscript entitled ‘I° Schema’ [First Scheme], Sraffa distinguishes three fields of work, namely, 1) a history of wage theories, 2) an analysis of how distribution is determined (still conceived on the basis of the marginalist theory) and 3) a number of specific issues such as trade unions and wage taxation (see D1/58: 1). A point that Sraffa focuses on in these manuscripts is the peculiarity of labour as a commodity. According to Sraffa, it stems from the fact that labour is a ‘composite’ commodity [D1/58: 3 and D1/68: 5–7], of which the amount sold by the worker may differ from that received by the buyer since the latter ascribes greater importance to efficiency and may try to increase the intensity of work for a given wage per unit of time. It also stems from the fact that the apparatus of marginal utility does not apply well to labour. A reason for this is that marginal utility increases until the wage rate does not keep the worker and his family alive.3 Sraffa thus observes [D1/67: 1; D1/62: 2] that Nansen’s choice to give foods to only a part of the Russian population during the famine of 1921 has a general theoretical importance since it shows that what is relevant in consumption is the dose of ‘food’ that assures a labourer’s survival. Sraffa also points out that a worker enjoyed subsistence as the owner of his labour force which is a perishable commodity, and not as a consumer (see also below, p. 9). Sraffa’s observation that what is relevant in consumption is the dose of ‘food’ able to keep a labourer alive is part of his criticism of the notion of utility and its applicability, which is a constant feature of his thought since his first systematic studies of economic theory—e.g. in his notes on Cannan’s Wealth of February 1923 [D1/67: 1 (13)], and in those on Marshall’s Principles of April 1923 [D1/2]. Thus, he stresses that the consumer has the choice of working eight hours a day or not at all, of buying a whole apartment and not one room in it, of eating soup and not one spoonful of it and one spoonful of another soup (see D1/66: 7). Sraffa also observes that the notion of consumer surplus cannot be applied to necessities, whose utility is infinite.4 Many other manuscripts possibly written by Sraffa in the years 1922–26 concentrate on the social nature of tastes and consumption and the difficulties of the ceteris paribus assumption in dealing with demand and supply curves (see e.g. D1/16–17 and D1/46).5 3. The debate on taxable income Sraffa’s reflections on the minimum wage in the years 1922–26 are thus interwoven with different strands of his initial research interests, namely, distribution theories and the labour market, ‘special issues’ such as taxation, the definition of net income and the problem of avoiding infinite values if a definite sum of utilities has to be reached as in consumer surplus (see e.g. D1/13 and parts of the manuscripts on Bernoulli in D1/11). These aspects are all present in the manuscript [D1/60] entitled Definizione di salario netto [Definition of net wages], in which Sraffa deeply analyses the definition of capital and income proposed by Cannan (1897), Fisher (1897A, 1897B, 1906), Marshall (1920 [1949]) and Nicholson (1891).6 Sraffa first emphasizes that the distinction between gross and net wages (namely, the wages net of subsistence) is present not only in Turgot and other classical economists, but also in Marshall and in the literature on taxable income,7 as well as in some contributions on the costs of production of men, like that of Benini (1924). For instance, with respect to Marshall, Sraffa annotates the page numbers where Marshall writes that ‘(i)n accordance with a suggestion made by Daniel Bernoulli, we may regard the satisfaction which a person derives from his income as commencing when he has enough to support life’ (Marshall [1920] 1949, p. 111). He also draws attention to the footnote in the same page of Marshall’s Principles which adds that ‘the systems of taxation which are now most widely prevalent follow generally on the lines of Bernoulli’s suggestion’. Finally, Sraffa annotates the page where Marshall observes that [p]roductive consumption, when employed as a technical term, is commonly defined as the use of wealth in the production of further wealth; and it should properly include not all the consumption of productive workers, but only that which is necessary for their efficiency. The term may perhaps be useful in studies of the accumulation of material wealth. But it is apt to mislead. For consumption is the end of production; and all wholesome consumption is productive of benefits, many of the most worthy of which do not directly contribute to the production of material wealth. (Marshall, [1920] 1949, p. 56, my emphasis)8 The ‘misleading idea’ stressed in this passage by Marshall is at the root of debates on the definition of net income and the practice followed in taxation to exempt what is needed ‘to support life’. As Sraffa himself outlines (see D1/12), the Report quoted also by Marshall and written by a Committee of economists such as Farr and Jevons suggests that in taxation we have to ‘(d)educt from the income … all the outgoings that belong to its production’. In fact, (t)hese outgoings are for the most part the sequels of production, wear and tear and depreciation involving cost of repairs, maintenance or replacement either of the source itself or of its value. By their deduction the source’s value considered as a capital or principal is maintained unimpaired, and the income left which would always bear the same relation to source that interest bears to principal, was called the source’s ‘interest value’ and was adopted as the common measure of assessment.9 In the specific case of labour, the Committee observes that ‘(t)he income from labour of men is subject to essential outgoings, costs of maintenance, depreciation, exhaustibility as the income from houses or from horse …. Like the labour of a horse … [the labour of men] undergoes a daily exhaustion of power that has to be supplied by food’. Hence, as noted by Edgeworth (1896, p. 374)—who, however, as Sraffa specified, criticized such a view10—we should ‘exclude from the income of the labourer the expenditure which is necessary for his efficiency’. In the literature on taxation, this need to deduct the subsistence wage from the income of labourers was deemed to be a condition for fair taxation (see e.g. Edgeworth, 1919, 1920) and not to clash with the principle of universality advanced by the eighteenth-century reformers against the ‘immunitas’ of the clergy and landlords (cf. Seligman, [1911] 1914, p. 226)—i.e. against what was the opposite of ‘the principle of faculty’ in taxation. Thus, Cohen Stuart ([1889] 1920, p. 445, my translation) observes that referring to the labourer’s ‘freeincome’ (namely, the income which is not necessary for ‘life preservation and lasting productivity’) follows J. S. Mill’s principle of the equal sacrifice in taxation11 since—as noted also by Loria (1920)—basic needs have an intensity that is immeasurably higher than others. Cohen Stuart also noted that the distinction made by the cameralists between the ‘abstract’ ability to pay taxes and the actual one (determined precisely by deducting from income what is necessary to survive) recalls ‘the Ricardian theory of net income’, which, like all Ricardo’s considerations, has ‘some importance for taxation’. On the other hand, according to Cohen Stuart, (e)ven when one criticizes this [Ricardian] theory ... it should be recognized that public authority must ensure with the utmost care that no one levies an amount of taxes greater than net income …. (Cohen Stuart, [1889] 1920, p. 455, my translation) as defined by Ricardo, namely net of the subsistence wage. The same view is found in Hobson (1919), according to whom only the surplus product, which is given by the elements of income that do not ‘dry up’ with taxation, is to be taxed.12 Indeed, with few exceptions such as that of Stamp (1922)—who maintained that the determination of taxable capacity depends on how taxation is used and the ‘prevailing national spirit’—there was a general consensus that the subsistence wage has to be exempted,13 albeit with different justifications (cf. also Musgrave, 1985). 4. Sraffa’s criticism of Fisher’s definition of net income However, as Sraffa notes (see D1/60: 13 [12]), even if Marshall and Pigou take ‘as the standard meaning of the term national dividend, that suggested by the practice of the British Income Tax Commissioners’, nevertheless, when defining net income, they include in it the part that is needed to maintain the labourer as an income producer (see e.g. Pigou [1924] 1946, vol. 3, p. 4).14 Marshall’s reasons in this respect are those advanced, for instance, by Smart and Masci, namely, that in this case we are dealing with the income of free workers and not of slaves, and that ‘all wholesome consumption’ is productive of benefits. So, according to Smart ([1899] 1916, p. 467, my translation), the search in the subsistence wage for the explanation and principle of remuneration means ... conceiving the factors of production as paid not according to the services that they offer, but to what it costs to generate and keep them alive. This is the principle of the work of slaves, not of free people: it is the principle according to which we feed horses so that they can pull the plough, and not the principle whereby we pay factors that assert their right to receive a share of a total product that is increasing. (my emphasis)15 On the other hand, when criticizing Loria’s (1909) distinction of the net product in two fragments—namely, the subsistence wage and the income exceeding the reintegration of the energies consumed by the workers in the production process—Masci (1913, p. 78, my translation) observes that ‘the true aim and function of subsistence is the life of the worker, the satisfaction of his most urgent needs: here subsistence accomplishes basically the same function as income, so that the former cannot be separated from the latter.’16 When dealing with these arguments, Sraffa focuses especially on Fisher’s works on capital and income, conceived by Fisher himself as a version of those of Walras ([1874] 1954) and Cannan (1897). Three elements characterize Fisher’s analysis. First, he criticizes Hermann’s and Schmoller’s notion of ‘earned income’ as what can be used or consumed without reducing capital, and states that it must instead be viewed as the income realized every year, that is, as the whole amount of services which are actually enjoyed independently of the fact whether the ‘original’ capital is consumed or not.17 Second, according to Fisher, savings must be exempted from taxation since otherwise they would be taxed twice, in the form of capital and incomes generated by it. He does not consider, however, any exemption for necessary expenses, unlike Einaudi (1916), who viewed them as a kind of saving. Third, Fisher (1896, 1904 and 1906) states that there is no sense in distinguishing, as done by Smith (and also by Marshall), wealth (the ‘natural objects owned by men and external to the owner’)18 and capital on the grounds that the latter is what produces an income for its owner, since any item of wealth generates a service enjoyed by the consumer.19 In this context in which the emphasis is on income as providing enjoyment for consumers, Fisher (1906) also faces the view outlined by Sraffa of the 1878 ‘Committee Report on a Measure of Value in Direct Taxation’. Asking himself if ‘fuel and labour’ are to be deducted ‘in the same way as raw materials’, Fisher writes: Some writers have gone so far as to claim that, just as the cost of feeding work animals must be deducted from the value of the work they do, so the cost of supporting labourers must be deducted from the value of their product. If this view were correct, it would seem that the labourer could not share at all in the distribution of the social income, since all that comes to him is deducted. (Fisher, 1906, p. 114) Indeed, in order to avoid ‘some of the manifest unfairness in the usual statistical comparisons which contrast a capitalist’s income with that of a labourer’, Fisher maintains that a subjective notion of income should be considered in which one’s net income is obtained by subtracting ‘from the subjective satisfactions the subjective efforts of attainment’ (Fisher, 1906, p. 171).20 It implies deducting from the gross income of a working man the sum ‘which the worker would be willing to sacrifice were it possible for him to avoid the disagreeable element’ (Fisher, 1906, p. 172).21 According to Fisher, however, no additional exemption for deterioration of value of the worker should be made, ‘unless … the man’s income is to be taxed in perpetuity’, namely, ‘after it had ceased and the man is dead’ (Fisher, 1897B, p. 535). Sraffa criticized these theses by directly tackling the arguments against the exemption of the subsistence wage. Sraffa first objects to Fisher’s idea of subtracting only the labour ‘disservices’ from gross income that, in this case, the ‘disservices’ stemming from consumption postponement should also be deducted, but they will exist ‘only if postponing consumption is a sacrifice that is made to receive interest, and not when savings would be made irrespective of the interest’ [D1/60: 9, my translation].22 Sraffa then notices [D1/60: 13] that by net profits and net rent we usually mean ‘what remains after deducting all the expenses that are required for production and to reintegrate the factor of production in its full capacity, identical to that prior to production’.23 Consequently, if ‘net’ with regard to wages means the same thing, all the expenses required for maintenance of the worker (expenses that are vital to production), as well as all those required for depreciation, namely, for reinstatement of the worker (the insurance share for ‘working life’, or, in other words, child care expenses), should be deducted. If the objection is raised that the expenses for maintenance of a capitalist should then be subtracted from net profits, the response would be: the life of the owner of the ‘worker’ (productive tool) is necessary, in a wage regime, in order for the worker to be able to produce, but the life of the capitalist is not necessary for capital to produce. The example of slavery clarifies this idea because, in this case, the owner of the worker is distinct from the labourer himself: the life of the former is not necessary for production and neither is that of the capitalist; the life of the latter is essential, as is the existence of capital.24 Sraffa also responds to the main objection to determining income as net of the subsistence wage, namely, that net national income should be equal to the mass of goods and services that provide a utility and that workers would also receive satisfaction from subsistence goods. In particular, Sraffa maintains that sum and differences of utility are absurd operations since total utility may be infinite,25 and that the refund of production expenses also gives a utility to the capitalist since capital, if it is not used, wears out rapidly, and the source of its renewal disappears, so that the depreciation of both capital and labour should be included or excluded in the national income. In Sraffa’s words: there are various responses: A) this objection implies that national income is conceived as a set of total utilities and not values{26}: hence, to obtain net income, all the disutilities derived from workers in their work should be deducted: the result would be the same as the one stated here on a logical level even if not with regard to quantities. B) the sum and differences of total utilities are absurd, impossible operations: total utilities are often infinite: if the disutilities, which are always finite, are subtracted from them, the result would always be infinite. What is the point in talking about changes in an income which is always infinite and therefore invariable. C) the refunding of production expenses provides an utility not only to the worker, but also to capital: indeed, if capital were not used in production, it would be used up very quickly and there would be no source to renew it. If it is true that a worker, even if he receives only a bare subsistence wage (expense repayment) has net satisfaction, (that in some way must be calculated in the national income) – a capitalist, regardless of the profit, also obtains significant benefit from the fact that, by using his capital in production, it is continuously reinstated, i.e. retained: this would be impossible, with the exception of a few cases, if it were not used. In short, irrespective of any surplus, both the worker and the capitalist receive, by producing, the satisfaction of being kept alive.27 In conclusion, in order to have a coherent definition of income, it must either include the depreciation (expenses) of capital and labour, or exclude both. Moreover, faced again with the argument that net income should be available for a man’s enjoyment, and that the necessary expenses for the reproduction of labourers differ from those of capital due to the fact that the ‘man-machine’ enjoys them, unlike what occurs for other kinds of capital, Sraffa specifies that this is untrue since ‘a free worker does not enjoy food in a way that is different from that of a slave or a horse’. Indeed, He does not enjoy these expenses as an ‘income subject’ consumer, but as the owner of his person since this is a perishable commodity …. The owner of a slave or of a horse enjoys (in addition to the net product of their work) the food they eat because it is a necessary condition if its capital is to continue to exist.28 Note that, contrary to what has been stated by Marshall ([1920], 1949, p. 504) and Masci, according to Sraffa there is no cynicism in treating workers as horses with respect to the subsistence wage. Indeed, the fact that the subsistence wage has to be treated as capital for free labourers as well ultimately reflects their ‘structural’ need to work and thus sell themselves in order to survive and reproduce. In this respect, they differ from slaves only because they manage their reproduction,29 both being, however, ‘the most costly of animals’ (cf. Nicholson, 1892, p. 478). 5. Subsistence and surplus wage From the above we can see that already in manuscripts probably written before or around 1927 Sraffa advanced a definition of net income as that obtained by deducting all the necessary expenses, including the subsistence wage, from the social product. He also specified that the distinction between the cost and surplus components of wages may give us a more concrete and precise definition of the ‘standard of comfort’ as what is ‘necessary to maintain the population in its present state of number and efficiency’ (D1/60: 13).30 In fact, if part of the necessities is not consumed, the worker wastes away, that is, he destroys his personal capital.31 Finally, probably following Marshall’s statement (1898, p. 53) that the problem to be examined in the theory of distribution is which laws govern the division of the surplus product, Sraffa pointed out in the same manuscript [see D1/60: 10–12] that a true problem of distribution occurs only for a surplus product that is distributed among the factors of production in proportions determined ‘by causes other than the mode of production’.32 According to Sraffa, this implicitly determines the two conditions to be satisfied so that a wage theory can be built: the first is that capital is invested by persons other than workers; the second, that the workers are the owners of their labour force since slaves do not receive any share of the surplus product (see also D3/12/7: 161 and D3/12/12: 28). In manuscripts written in the years 1927–28 and in the Lectures on the Theories of Value and Distribution of 1928–31, Sraffa deeply examines these points by rediscovering (see Garegnani, 2005) the classical or surplus approach and overcoming his previous theoretical position of 1925–26.33 In particular, he further reflects on the notions of cost and the subsistence wage. By analysing the change that has occurred during the history of economic thought in the concept of cost from that in terms of ‘the means necessary to enable production to be made’ to that of ‘the incentive required to induce somebody to overcome the sacrifice involved in production’ [D3/12/4: 7], Sraffa criticized this latter Marshallian notion of subjective costs (see Gerhke and Kurz, 2006). With regard to the wage of a slave worker, Sraffa thus observes [D3/12/7: 106] that it is not necessary to ‘induce’ him to work (since he is not free), but ‘to enable’ him to work. This distinction between what is necessary to enable or to induce the labourer to work applies, according to Sraffa, also to the necessities of a free worker (see D3/12/9: 31; D3/12/9: 89; D3/12/10: 61 and 97), and it leads to conceiving them as a part of capital, as Sraffa specifies in his 1928–31 Lectures on value and distribution [D2/4: 24]: (t)his view [of wages as necessary subsistence for workers and not as an inducement to the ‘sacrifice’ of working] leads to a conception of wages and profits as two things of a radically different nature. Wages are a stock of goods that exists before production and which is destroyed during the productive process: they come thus to be identified with capital or at least with an important part of capital. Profits (and rent of course) are a part of the product and precisely the excess of the product over the initial stock.34 However, as mentioned above, in Sraffa’s manuscripts there are also some hints regarding the relation between the subsistence and surplus wage and a definition of the subsistence wage itself (see also Pivetti, 2000; Picchio, 2011). As is known, the classical economists and Marx admitted that workers may be strong enough in wage bargaining to obtain a share of the surplus product.35Smith ([1776] 1976, II, III, p. 354) for instance writes that wages may happen to pay taxes and sustain unproductive labourers,36 and Ricardo (1951–73, I, XXVI, p. 348, note) states that more is generally allotted to the labourer under the name of wages, than the absolutely necessary expenses of production. In that case a part of the net produce of the country is received by the labourer, and may be saved or expended by him ….37 This possibility of a wage that is higher than what is needed to keep workers ‘alive and efficient’38 is, on the other hand, implicit in Smith’s and Ricardo’s definition of the subsistence wage as historically determined. The subsistence wage is not in fact viewed by them as the amount of goods physiologically necessary for the survival and reproduction of labourers. As Ricardo points out, the minimum or necessary price of labour must also include the goods and services which the habits of a country—that operate as a ‘second nature’ (cf. Torrens, 1815, p. 63)—make necessary for the reproduction of efficient labourers. These commodities are actually something without which «a creditable day-labourer would be ashamed to appear in public» (cf. Smith, [1776] 1976, V, II, p. 399), and which assure the maintenance and reproduction of workers according to the ‘degree of civilization of a country … the conditions under which, and consequently … the habits and degree of comforts in which, the class of free labourers has been formed’ (Marx, 1961–63, vol. 1, p. 171). As stated also by Sraffa in a letter to Garegnani dated 13-3-62 (D3/12/111: 149, my translation):39 First of all, ‘subsistence has never meant pure ‘physiological necessity’ (whatever that means) but always includes social and historical or habitual necessity. This is clear, if only because when wages are reduced, people will often give up something physiologically necessary before giving up a ‘superfluous’ thing like alcohol, smoking, etc. And then when a ‘standard’ level has prevailed for a certain time, it becomes necessary—if you want the result.40 What concretely is the ‘bundle’ of necessaries that assures, as written by Marx, the formation, reparation, and reintegration of the labourer is of course not easy to determine. One reason is that, if commodities have substitutes, there may be, Sraffa notes, ‘an infinite number of combinations of the different commodities, which satisfy the condition of maintaining life and efficiency of producers’ (D3/12/3: 44). Nevertheless, Sraffa also notes that ‘the difficulties of distinguishing in the total money cost of a thing what is real cost and what is surplus may be very great in practice, but it is not greater than other similar distinctions that are accepted in every economic theory with regard to rent, interest, etc.’ (D3/12/42: 35). The necessaries appear in fact to be settled in the hierarchy of needs and ways to meet them, with the related distinction between necessary and luxury goods (including the luxuries of the poor mentioned by Adam Smith), which are inherited from the past and embodied in the habits of consumption, social norms and specific institutional arrangements of a given society. Their content will usually change only after a significant change in the wage goods prices occurs, or as a consequence of consumption innovations (especially when the prices of the new goods become relatively low), or when the wage rate remains for a sufficient period of time above the subsistence level.41 Sraffa also reflects on the ‘practical’ (or ‘statistical’) difficulties of determining the subsistence wage and separating it from the surplus wage in the period when Production of Commodities by Means of Commodities (PCMC) was published, pointing out again that these difficulties are ‘not insurmountable’ or ‘greater than the existing one of deciding whether ‘working clothes’ are part of wages—& how to sort out shares if in quality they are better than strictly necessary’ (D3/12/52: 8). It is therefore not surprising that in PCMC Sraffa writes that ‘it would be appropriate … to separate the two component parts of the wage and regard only the “surplus” part as variable’, or consider the influence of necessaries on prices and profits ‘by setting a limit below which the wage cannot fall’ (Sraffa, 1960, pp. 9–10)—a limit provided precisely by the price of the bundle of necessaries.42 6. Some final remarks for tax policy The above analysis restates Ricardo’s view (cf. Ricardo, 1951–73: vol. I, XXVI, pp. 347–48) that taxable capacity is given by net income (or surplus product), namely, that of the three parts into which the product is divided—wages, profits and rents—‘from the last two portions only … any deduction can be made for taxes, or for savings: the former, if moderate, constituting always the necessary expenses of production’. As Sraffa noted in November 1927 in a manuscript entitled ‘Surplus theory and taxation’ [D3/12/11: 56], this is indeed the view that is generally followed in public finance which ‘acts on the theory of the surplus, by necessity’: Surplus theory and taxation How public finance (the only practical part of economics) acts on the theory of the surplus, by necessity: 1) exemption of minimum wages (Report of B{ritish} Ass{ociation}1878) 2) exemption of saving (Mill’s theorem)43 (surplus saved) 3) direct taxation is harmless, indirect disastrous in its effects on value (all the theory in this sense is based on physiocratic and Ricardian net product) where the harmless effect of direct taxation seems to refer to the fact that both savings and the necessary expenses of production are to be exempted and only rents or other ‘surpluses’ should be taxed.44 Here Sraffa probably had in mind the literature on taxation in which taxable income is viewed as the value of goods and services that can be freely consumed without drying up their sources, comprising the source of labour which undergoes an ‘exhaustion of power’ that must be replaced (see above, Section 3 and note 17). This need to exempt a minimum for the maintenance of labourers and/or introduce (cf. for instance, Kaldor, 1956, p. 28; Stiglitz, 2001) allowances and deductions for lower incomes is still recognized in fiscal theory and followed in taxation, where, as written by Kaldor (1980, p. 2), taxable capacity is deemed to be equal to that part of income that is in excess of ‘material needs’. As in the classical economists (cf. Smith [1776] 1976, I, VIII, p. 86; IV, V, p. 49), one reason for this is that stated for instance by Hayek (1960, p. 285), who observes that ‘(t)he necessity of some such arrangement in an industrial society is unquestioned—be it only in the interest of those who require protection against acts of desperation on the part of the needy’, thus recalling that wages cannot fall below the subsistence level if an orderly reproduction of the economy is to be guaranteed. There are, however, several problems to be solved in following the practice of exempting subsistence wage from taxation, and they may find different solutions according to the starting theory of distribution, and the kind and extent of State intervention in the labour market and in providing social goods and benefits. In concluding this work, I will briefly outline some of these difficulties and the related theme of an operational determination of the subsistence wage, which, as seen above, Sraffa refers to in his manuscripts. The first issue concerns who should in actual fact be exempted from taxation. Cohen Stuart restricts the problem to unskilled and skilled workers by observing that those who are ‘forced by their profession to dress as a gentleman, to keep a butler, to give a party and so on would use their income in the same way even if their profession did not impose this obligation on them’ (Cohen Stuart, [1889] 1920, p. 455, my translation)—thus stressing that in this case these expenses are not really made by necessity.45 The second issue regards the level of income to be exempted and brings us back to the notion of subsistence and its operational determination. On the whole, the bundle of necessary or basic goods must include ‘some definite quantity and quality of house accommodation, of medical care, of education, of food, of leisure, of the apparatus of sanitary convenience and safety when work is carried on, and so on’ (Pigou, [1924] 1946, p. 724), and although absolute in a certain period, it will probably vary according to per capita income (cf. also Townsend, 1970, 1985).46 It is not, however, a ‘living income’, that is, the necessary expenses of a worker must refer to a man with ‘average good fortune’ in the matter of sickness and an average number of children (cf. Pigou, [1924] 1946, p. 558). Moreover, these expenses will vary during the labourer’s lifetime, thus implying savings and dissavings in certain periods of life (see e.g. Bowley, 1913; Engel, 1866) according to the length of time in which the ‘man-machine’ is built up, used and ‘scrapped’. Finally, the corresponding average wage rate will be greater or lower according to the required technological knowledge and the contribution to the family income of female earnings (net of costs of the services to be bought in the absence of household work) as influenced by demographic and social factors, since it is the family income that is relevant to the reproduction of labourers.47 As far as an operational definition for taxation is concerned, as a first step the method usually used to determine poverty lines can be followed in which necessary goods stem (in the words of A. Giffen) from ‘a wise calculation’ taking into account wage agreements and the activities of the Courts, as well as the actual patterns of consumption of society in a certain period of time48—and not, as sometime stated, from an arbitrary value judgment of a ‘spectator’. In Rowntree (1901) or Orshansky (1965), for instance, poverty lines are specified by means of E=(1+h)PX* where X* is a targeted consumption basket, P the vector price, and h a percentage that takes into account the expenses for those needs (such as transportation and personal care) which are not considered in the basket X* since their determination is less immediate and easy.49 Unlike other approaches,50X* and h are derived by means of a direct enquiry of the expenditures of the workers’ families, or by the spending pattern of all the families below the median expenditure level (cf. Brown, 1994). However, a certain consumption level could also be maintained by dissaving or borrowing if the income Y is lower than (1+h)PX* and, as said above, the minimum wage does not exclude the need for saving. To consider these aspects, reference can be made to a certain income threshold Y*, achieved by choosing an appropriate value h* (see e.g. Atkinson, 1989). The amount of Y* may vary according to the purposes in mind. For example, if it refers to what is needed to assure subsistence, and not to the mere survival of the poorest people as in some programmes against famine and destitution,51 it will include what is necessary to participate in society, because social beings are expected to perform a socially demanding role as workers, citizens, parents and friends, that is, they must have the ability to function rather than only to survive.52 Furthermore, since the kind and amount of necessities to perform certain activities will change over time (usually in line with per capita income), this objective and ‘absolute’ approach does not imply fixed subsistence lines and does not neglect the fact that our sense of privation is socially relative and stronger if there is an increase in the standard of living of the upper classes (cf. Marx ([1884] 1978, p. 33).53 But two further points must be considered when following Sraffa’s suggestions on wage tax policy. First, even when an amount of income Y* is exempted from taxation, there may be the need for other forms of State intervention. This is the case for actual wages falling on average below subsistence levels (as possibly, for instance, in a ‘retrograde state’ of the economy), or tax exemptions not sufficiently taking into account that the subsistence wage is one that is able to reproduce an efficient worker, thus changing according to the kind and intensity of work that is performed.54 Second, when determining the amount of Y*, it should be considered that some of the wage goods are provided by social security55 and that State intervention redistributes income and wealth and relieves poverty through subsidies against unemployment or minimum pensions. A way to face the first point is to introduce a minimum wage in any industry and occupation, as done for instance in the UK for the greatest part of the twentieth century by the Boards of Trade for the various industries. Also in the UK, however, this approach was accompanied by the proposal for a general minimum wage in order to fight the sweating system, incentivize more advanced techniques and assure a minimum standard of living for the population.56 There is more need for a measure of this kind the greater the drop in union coverage and the higher the number of low-paid workers and the level of unemployment. With respect to the second point, the set of goods obtained by workers includes the amount of social goods provided by the State, with the net social wage (namely, the value of these social goods net of taxation and social contributions) usually being higher for the poorest sectors of population due to tax exemptions and subsidies.57 Consequently, wages net of taxes have to be greater where the amount of wage goods supplied by the State is lower, and the State policy itself becomes a terrain of conflict between classes (cf. Cesaratto, 2010; Gough, 1979) with respect to the amount of family needs covered by public transfers and goods, as well as the ways they are financed. However, as Sraffa noted, the notion of necessity has only a definite meaning from a given point of view, which must be explicitly stated, & then adhered to consistently. … The standpoint of capitalist society itself, is that of the ruling class … [D2/12/7: 161] and this standpoint tends to put a limit to the amount of basic needs and social rights provided by the State. It implies that the reference in Sraffa (1960) to a subsistence wage should not be used to directly sustain ideas such as that of a universal basic income (see van Parjis and Vanderborgth, 2013), as suggested by Chiodi and Ditta (2013).58 Such proposals should find their ground in welfarist considerations that go beyond the standpoint of capitalist society. As is hopefully clear from the analysis carried out so far, Sraffa’s thoughts on subsistence wage and taxable income may nevertheless provide the basis for a criticism of some recent tendencies in wage and fiscal policies aimed at reducing tax progressivity and increasing the ‘flexibility’ of the labour market—with their results in terms of greater wage variability and even sometimes a fall on average in net real wages. It also provides insights into tax policy and the determination of minimum wages which seem especially relevant with regard to the experience of the main industrialised countries in the present context (see Piketty, 2013) of increasing income and wealth inequality. In particular, two previously analysed points can be outlined. First, Sraffa’s revival of the surplus approach and his emphasis on taxable income as net of the subsistence wage can give a solid basis to the fiscal practice of introducing tax allowances and deductions for lower incomes—a practice which is to be followed when an orderly reproduction of the economy must be guaranteed. Second, the amount of wages to be exempted from taxation should consider the whole set of factors that shape the subsistence wages of the different groups of workers. Of course, the practical solution to this ought to be different in the various countries according to the historical evolution of the relations between labour and capital and the State provision of social goods that can be found in each of them. A simple (though imperfect) way of tackling it may be, however, to exempt from taxation a family income (net of workers’ social contributions) which is equal to what is needed to guarantee the reproduction of an average unskilled worker,59 while conceding to the other workers tax allowances based on the intensity and skill of the work performed. The possible drop in fiscal revenues that this solution implies could be compensated for by greater tax progressivity as well as by overcoming the preferential tax regime for capital income and wealth that has become almost natural in recent decades and has led to the paradox that the tax burden falls mostly on labour. I wish to thank two referees for their suggestions on a previous version of this paper. I also wish to thank Professor Lord John Eatwell for granting me permission to publish from Sraffa’s unpublished archive. Bibliography Atkinson , A. B . 1989 . Poverty and Social Security , New York , Harvester Wheatsheaf Atkinson , A. B . 1995 . Public Economics in Action: The Basic Income/Flat Tax Proposal , Oxford , Clarendon Press Barbosa de Melo , F. L. , Figueiredo , A. , Mineiro , A. S. and Arbulu Mendonca , S. E . 2012 . Rescuing the minimum wage as a tool for development in Brazil , in International Journal of Labour Research , 4, 1, ‘Social Justice and Growth: The Role of the Minimum Wage’, ILO , Geneve, pp. 27–44 Barton , J. [ 1817 ] 1934 . Observations on the Circumstances which Influence the Condition of the Labouring Classes of Society , Baltimore , Johns Hopkins Press [SL: 8095] Battistella , C . 1913 . Il Concetto di Reddito in Economia, in Finanza e nel Diritto Finanziario , Rovigo , Tipografia Sociale Benini , R . 1924 . Da Malthus e Ricardo a Vilfredo Pareto, Traccie di una via maestra , Giornale degli Economisti e Rivista di Statistica , vol. 39 , no. 1–2 , 90 – 102 Beveridge , Sir W. (Lord) 1942 . Social Insurance and Allied Services , London , HMSO Bowley , A. L . 1913 . Working class households in Reading , Journal of the Royal Statistical Society , vol. 76 , 672 – 701 Google Scholar CrossRef Search ADS Bowley , A. L . 1922 . The definition of national income , Economic Journal , vol. 32 , no. 125 , 1 – 11 Google Scholar CrossRef Search ADS Brown , C . 1994 . American Standard of Living , Oxford , Basil Blackwell Cannan , E . 1897 . What is capital ?, Economic Journal , vol. 7 , 278 – 84 Google Scholar CrossRef Search ADS Cannan , E. [ 1917 ] 1967 . A History of the Theories of Production and Distribution in English Political Economy from 1776 to 1848 ( 3rd ed .), A. M. Kelley reprints , New York [SL: 7929, 1st ed.] Card , D. and Krueger , A. B . 1995 . Myth and Measurement. The New Economics of the Minimum Wages , Princeton, NJ , Princeton University Press Cesaratto , S . 2010 . The classical surplus approach and the theory of the welfare state and public pensions , in Chiodi , G. and Ditta , L ., Sraffa or an Alternative Economics , London , Macmillan Google Scholar CrossRef Search ADS Chapman , S. J . 1913 . The utility of income and progressive taxation , Economic Journal , vol. 23 , no. 89 , 25 – 35 Google Scholar CrossRef Search ADS Chiodi , G. and Ditta , L . 2013 . Sraffa and Keynes: two ways of making a ‘revolution’ in economic theory , in Levrero , E. S. , Palumbo , A. and Stirati , A ., Sraffa and the Reconstruction of Economic Theory: Volume One , London , Macmillan Google Scholar CrossRef Search ADS Citro , C. F. and Michael , R. T . (eds.) 1995 . Measuring Poverty. A New Approach , Washington, DC , National Academies Press Cohen Stuart , A. J. [ 1889 ] 1920 . Contributo alla Teoria Progressiva dell’Imposta sul Reddito, Biblioteca dell’Economista, traduzione di Achille Loria , vol. 15 , Unione tipografico-editrice torinese , Torino Edgeworth , Y. F . 1896 . Income , in Palgrave , I . (ed.), Dictionary of Political Economy , vol. 2 , London , Macmillan Edgeworth , Y. F . 1919 . Methods of graduating taxes on income and capital , Economic Journal , vol. 29 , no. 114 , 138 – 53 Google Scholar CrossRef Search ADS Edgeworth , Y. F . 1920 . La Teoria Pura dell’Imposta , in Biblioteca dell’Economista , vol. 16 , Unione tipografico-editrice torinese , Torino Einaudi , L . 1912 . Intorno al Concetto di Reddito Imponibile e di un Sistema di Imposte sul Reddito Consumato , Memorie della Accademia delle Scienze di Torino , Torino [SL: 8454] Einaudi , L . 1916 . Corso di Scienze delle Finanze , Terza edizione, Edizione della Rivista La Riforma Sociale , Torino [SL: 1713] Engel , E . 1866 . Der Preis der Arbeiter , Berlin , Luderitz Eurostat 2014 . Living Conditions in Europe , Luxembourg , Publications Office of the European Union Fisher , I . 1896 . What is capital ?, Economic Journal , vol. 6 , no. 24 , 509 – 34 Google Scholar CrossRef Search ADS Fisher , I . 1897A . Senses of capital , Economic Journal , vol. 7 , no. 26 , 199 – 213 Google Scholar CrossRef Search ADS Fisher , I . 1897B . The role of capital in economic theory , Economic Journal , vol. 7 , no. 28 , 511 – 37 Google Scholar CrossRef Search ADS Fisher , I . 1904 . Precedents for defining capital , Quarterly Journal of Economics , vol. 18 , no. 3 , 386 – 408 Google Scholar CrossRef Search ADS Fisher , I . 1906 . The Nature of Capital and Income , New York , Macmillan [SL: 2690] Fuchs , V . 1965 . Towards a theory of poverty , in Task Force on Economic Growth and Opportunity, The Concept of Poverty , Washington, DC , Chamber of Commerce of the United States of America Garegnani , P . 2005 . On a turning point in Sraffa’s theoretical and interpretative position in the late 1920s , European Journal of the History of Economic Thought , vol. 12 , no. 3 , 453 – 92 Google Scholar CrossRef Search ADS Gehrke , C . 2016 . Subsistence-cum-surplus wages versus gross wages: a note , in Freni , G. , Kurz , H. D. , Lavezzi , A. M. and Signorino , R . (eds.), Economic Theory and its History. Essays in Honour of Neri Salvadori , London and New York , Routledge Gehrke , C. and Kurz , H. D . 2006 . Sraffa on von Bortkiewicz: reconstructing the classical theory of value and distribution , History of Political Economy , vol. 38 , no. 1 , 91 – 149 Google Scholar CrossRef Search ADS Gough , I . 1979 . The Political Economy of the Welfare State , London , Macmillan Google Scholar CrossRef Search ADS Hayek , F . 1960 . The Constitution of Liberty , Chicago , University of Chicago Press Hildebrand , G. H . 1967 . Poverty, Income Maintenance and the Negative Income Tax , ILR Paperback, 1, April Hobson , J. A . 1919 . Taxation in the New State , London , Methuen Istat 2009 . La Misura della Povertà Assoluta. Metodi e Norme , Roma , Istituto Nazionale di Statistica Kaldor , N . 1956 . An Expenditure Tax , London , George Allen and Unwin Kaldor , N . 1980 . Reports on Taxation , I, in Kaldor , N ., Collected Essays , vol. 7 , London , Duckworth Kurz , H. D . 2014 . Sraffa’s equations ‘Unveiled’? A comment on Gilibert , in Kurz , H. D. and Salvadori , N . (eds.), Revisiting Classical Economics: Studies in Long-Period Analysis , London , Routledge Levine , D. P. and Abu Turab Rizvi , S. 2005 . Poverty, Work and Freedom: Political Economy and the Moral Order , Cambridge , Cambridge University Press Google Scholar CrossRef Search ADS Levrero , E. S . 2000 . Crescita e distribuzione in Von Neumann e l’analisi di Sraffa , in Pivetti , M . (ed), Piero Sraffa: Contributi per una biografia intellettuale , Roma , Carocci Levrero , E. S . 2012 . Four Lectures on Wages and the Labour Market , Roma , Aracne Levrero , E. S . 2013 . Marx on absolute and relative wages and the modern theory of distribution , Review of Political Economy , vol. 25 , no. 1 , 91 – 116 Google Scholar CrossRef Search ADS Loria , A . 1909 . La Sintesi Economica: Studio sulle Leggi del Reddito , Torino , Bocca Loria , A . 1920 . Prefazione a Cohen Stuart, Contributo alla Teoria Progressiva dell’Imposta sul Reddito, Biblioteca dell’Economista, traduzione di Achille Loria , vol. 15 , Unione tipografico-editrice torinese , Torino Marshall , A. [ 1920 ] 1949 . Principles of Economics ( 8th ed .), London , Macmillan [1922 SL: 2591] Marshall , A . 1898 . Distribution and exchange , Economic Journal , vol. 8 , no. 29 , 37 – 59 Google Scholar CrossRef Search ADS Marshall , T. H . 1950 . Citizenship and Social Class , Cambridge , Cambridge University Press Marx , K . [ 1867-94 ] 1961-63 . Capital: A Critique of Political Economy, Volumes I-III , Foreign Languages Publishing House , Moscow Marx , K. [ 1884 ] 1978 . Wage, Labour and Capital , Peking , Foreign Languages Press Masci , G . 1913 . Il Concetto e la Definizione del Reddito , Napoli , Pierro Meade , J. E . 1972 . Poverty in the welfare state , Oxford Economic Papers , vol. 24 , 289 – 326 Google Scholar CrossRef Search ADS Musgrave , R. A . 1985 . Introduction , in Auerbach , A. J. and Feldstein , M ., Handbook of Public Economics , vol. 1 , Amsterdam, North Holland Nicholson , J. S . 1891 . The living capital of the United Kingdom , Economic Journal , vol. 1 , no. 1 , 95 – 107 Google Scholar CrossRef Search ADS Nicholson , J. S . 1892 . Capital and labour: their relative strength , Economic Journal , vol. 2 , no. 7 , 478 – 90 Google Scholar CrossRef Search ADS Orshansky , M . 1965 . Counting the poor: another look at the poverty profile , Social Security Bulletin , vol. 28 , 3 – 29 Pantaleoni , M. [ 1889 ] 1911 . Principi di Economia Politica , G. Barbera editore, Firenze [SL: 2302, 2nd ed., annotated by Sraffa] Pantaleoni , M. [1882] 1958 . Teoria della Traslazione dei Tributi , Milano , Giuffrè [SL: 876] Picchio , A . 2011 . Hay, carrots, bread and roses: net and subsistence wages in Sraffa’s papers , in Ciccone , R. , Gehrke , C. and Mongiovi , G . (eds.), Sraffa and Modern Economics , London , Routledge Pigou , A. C . [ 1924 ] 1946 . The Economics of Welfare , London , Macmillan [SL: 2597, 1st ed. (1921) annotated by Sraffa] Piketty , T . 2013 . Le Capital au XXIeSiecle , Paris , Edition du Seuil Pivetti , M . 2000 . Il concetto di salario come costo e sovrappiù e le sue implicazioni di politica economica , in Pivetti , M . (ed.), Piero Sraffa , Roma , Carocci Rathborne , E. F . 1920 . The new South Wales scheme for the grading of wages according to family needs , Economic Journal , vol. 30 , no. 120 , 550 – 53 Google Scholar CrossRef Search ADS Ricardo , D . 1951–73 . The Works and Correspondence of David Ricardo , edited by Piero Sraffa , Cambridge , Cambridge University Press Ricca Salerno , G . 1900 . La Teoria del Salario nella Storia delle Dottrine e dei Fatti Economici , Palermo , A. Reber Ricci , U . 1913 . Che cosa è il reddito , in Giornale degli Economisti e Rivista di Statistica , serie terza, 24, vol. 47 , no. 2 , Roma Athenaeum Rowntree , B. S . 1901 . Poverty: A Study of Town Life , London , Macmillan Seligman , E. R. A. [ 1911 ] 1914 . The Income Tax: a Study of the History, Theory and Practice of Income Tax at Home and Abroad , New York , Macmillan Sen , A. K . 1979 . Issues in the measurement of poverty , Scandinavian Journal of Economics , vol. 81 , 285 – 307 Google Scholar CrossRef Search ADS Sen , A. K . 1983 . Poor relatively speaking , Oxford Economic Papers , vol. 35 , no. 2 , 153 – 69 Google Scholar CrossRef Search ADS Sen , A. K . 1984 . The living standard , Oxford Economic Papers , vol. 36 , 74 – 90 Google Scholar CrossRef Search ADS Shaikh , A. and Tonak , E. A . 1987 . The welfare state and the myth of the social wage , in Cherry , R , et al. (eds.), The Imperiled Economy , vol. 1 , New York , URPE Smart , W . 1891 . An Introduction to the Theory of Value: on the Lines of Menger, Wieser, and Bohm-Bawerk , London , Macmillan [SL: 2306] Smart , W. [ 1899 ] 1916 . La Distribuzione del Reddito , in Biblioteca degli Economisti , Quinta serie, vol. 3 (Italian translation of The Distribution of Income, London , Macmillan ) Smith , A. [ 1776 ] 1976 . An Inquiry into the Nature and Causes of the Wealth of Nations , Chicago , Chicago University Press Google Scholar CrossRef Search ADS Sraffa , P . 1926 . The laws of return under competitive conditions , Economic Journal , vol. 36 , no. 144 , 535 – 50 Google Scholar CrossRef Search ADS Sraffa , P . 1960 . Production of Commodities by Means of Commodities , Cambridge , Cambridge University Press Stamp , J . 1922 . Wealth and Taxable Capacity , London , P. S. King & Son Stigler , G. J . 1946 . The economics of minimum wage legislation , American Economic Review , vol. 36 , no. 3 , 358 – 65 Stiglitz , J . 2001 . Economics of the Public Sector , New York , Norton Streeten , P. and Burki , S. J . 1978 . Basic needs: some issues , World Development , vol. 6 , 3 , 411 – 421 . Google Scholar CrossRef Search ADS Torrens , R . 1815 . An Essay on the External Corn Trade , London , J. Hatchard , [SL: 1393] Townsend , P . (ed.) 1970 . The Concept of Poverty , London , Heineman Townsend , P . 1985 . A sociological approach to the measurement of poverty: a rejoinder to Prof. A. Sen , Oxford Economic Papers , vol. 37 , 659 – 68 Google Scholar CrossRef Search ADS Van Parijs , P. and Vanderborgth , Y. 2013 . Il Reddito Minimo Universale , Milano , Università Bocconi Editore Van Pragg , B. , Goedhart , T. and Kapteyn , A . 1980 . The poverty line , Review of Economics and Statistics , vol. 62 , 461 – 85 Google Scholar CrossRef Search ADS Walras , L. [ 1874 ] 1954 . Elements of Pure Economics , London , Irwin Williams , G . 1936 . The State and the Standard of Living , London , P. S. King & Son Ltd Wueller , P. H . 1938 . Concepts of taxable income: vol. 1, The German Contribution; vol. 2, The American Contribution; vol. 3, The Italian contribution , Political Science Quarterly , New York, vol. 3 , no. 1 and 2 ; 1939, vol. 4, n. 4 Footnotes 1 I refer to the manuscripts D1/58–D1/68 in Miscellaneous Theory 1928–1931. They were classified as belonging to the period 1928–31 probably due to the dating of the initial two manuscripts D1/54–55 of Miscellaneous Theory 1928–1931, the fact that they contain writings on themes which to some extent are linked to Sraffa’s Lectures on the Theories of Value held in Cambridge in the years 1928–31 and the fact that they are all in the same brown parcel which also includes papers on capital theory, Keynes’s Treatise and the Symposium. The manuscripts D1/58–D1/68 can be traced back, however, to the period up to 1927. To a large extent, they appear to reflect Sraffa’s early interest in labour economics (as testified, for instance, by his participation in the activities of the Labour Research Department when he was staying in London in 1921), and his first studies of Marshall’s theory of value. Of course, we cannot be completely sure of the dates. Yet, both the content of these manuscripts and the fact that they are written in Italian suggest that they may not belong to the years 1928–31. There are also some explicit hints in this direction. For instance, there are manuscripts written by Sraffa when he was Director of the Labour Office of Milan (he resigned from that post in December 1922), a manuscript on Cannan’s Wealth dated February 1923 and a short summary and small selection of articles published in 1921 in the Communist Review and other journals. There are also some phrases that were written in small strips of the Corriere della Sera newspaper dated 1923 and 1924 and a list of references to the Collegio dei Cambi of Perugia probably written when he was teaching here in the years 1923–25. 2 Marshall ([1920], 1949, pp. 466–81) refers to the fact that the worker sells his work but he himself has no price, that labour is a perishable commodity and its sellers are often at a disadvantage in wage bargaining and that a long period of time is required to provide additional labour supplies, especially specialized abilities. 3 In this respect, Sraffa (his emphasis and question mark) quotes (D1/68: 7) Marshall’s phrase that ‘[i]t may be noticed here, though the fact is of but little practical importance, (?) that a small quantity of a commodity may be insufficient to meet a certain special want; and then there will be a more than proportionate increase of pleasure when the consumer gets enough of it to enable him to attain the desired end’ (Marshall, [1920] 1949, p. 79, n. 2). Sraffa observes that Marshall refers to the example of ‘little practical importance’ of wallpaper, but that this ‘particular’ case of a ‘more than proportional increase of pleasure’ includes nearly all the wage (i.e. the subsistence wage). 4 So Sraffa writes (D1/2: 4], my translation): ‘p. 133 The ‘necessary’ goods seemed to have been left out of the theory of consumer surplus: here, Marshall affirms, however, that they are the main case. In order to avoid the difficulty arising from the fact that the utility of necessities is infinite, Marshall suggests (in the footnote) that we consider only that part of the commodity that exceeds ‘what is strictly necessary to live’ (See Nicholson’s answer, Principles, p. 53.) So, in actual fact, the theory does not apply to necessities!’ [p. 133 Le cose ‘necessarie’ sembravano eccettuate dalla teoria del consumers’ surplus: qui M{arshall} afferma che esse sono il caso principale. Per evitare la difficoltà che sorge dal fatto che l’utilità delle cose necessarie è infinita, M{arshall} propone (in nota) di considerare solo quella parte della merce che supera ‘lo stretto necessario per vivere’. (V{edi} risposta di Nicholson, Principles, p. 53). Ma dunque effettivamente la teoria non si applica alle cose necessarie!] 5 According to Sraffa, due to the indivisibility of doses in consumption, the apparatus of marginal utility would apply to the demand of one man for many commodities (namely, to the utility of wealth in general), or to the demand of a large number of agents for one commodity (when indivisibility averages out). But these are cases in which ‘Pigou acknowledges that you cannot get by with the condition “other things being equal”’ [Pigou riconosce che con l’other things being equal’ non si può cavarsela] and the consumer surplus cannot be calculated (see D1/66: 7). 6 The dating of this manuscript is definitely after 1923 (Sraffa quotes an article written in 1924) and should probably be placed between 1925 and 1927. A reference to the work’s subject ‘Definizione di salario netto’ [Definition of net wage] is included in a list written in Italian probably for the years 1923–24 (see D1/62: 15). 7 Sraffa refers to the works of Battistella (1913), Edgeworth (1896), Einaudi (1916), Seligman ([1911] 1914) and Stamp (1922). See below, pp. 5–8. 8 Note also that in the copy of Marshall’s Principles (1922, 8th ed.) in Sraffa’s library [2591], Sraffa signed note 1 on page 66 and page 70 where Marshall writes that necessaries should be included in the means of production. 9 See ‘Report of the Meeting of the Committee of the British Association on a Common Measure of Value in Direct Taxation’, 1878, pp. 221–22. Sraffa also quotes this passage in manuscript D3/12/2: 23–24 dated 15/2/46 and writes in the top left of the page ‘(From folder: Physical real costs).’ 10 Edgeworth maintained that ‘there is a difference, that the cart-horse’s food and gear are not “goods of the first order”, objects of human consumption’. 11 Note that from a utilitarian standpoint, Edgeworth (1920) showed that this principle may imply proportional or progressive tax rates according to the behaviour of marginal utility when income changes. When, as in Chapman (1913, pp. 33–34, quoted by Sraffa), a difficulty was stressed in comparing the utilities of different consumers, tax progressivity was proposed again on the grounds that ‘the wants satisfied by the earlier increments to income are usually of more importance socially than the wants satisfied by later increments to income, whether the satisfaction of the former causes more utility or not’, because some confiscation of income to the poorer man is the more likely ‘to cause him deprivation of comforts which add to efficiency (meaning the social value of his life)’. 12 The idea of income sources that should not be dried up is also present in Smith and Marx according to Sraffa (see the manuscript D1/22). 13 Among others, Sraffa mentions Seligman ([1911] 1914) and Einaudi (1916), the latter conceiving necessary or ‘primary’ consumption as personal saving to be exempted from taxation both because savings would have in general to be exempted and because workers would not otherwise be productive or able to attain pleasure (see Einaudi, 1916, p. 205). Einaudi, however, stated that other expenses, even of the richer classes, should also be exempted if ‘productive’—a view criticized by Ricci (1913, p. 322, my translation), who maintains that ‘going to the cinema, going out for the day on a Sunday, reading newspapers and informative books’ may certainly ‘make work more productive’, but then all these ‘unnecessary’ expenses ‘would become savings’ to be exempted. 14 There are, however, some ambiguities in both Marshall and Pigou in this respect. Thus, Pigou ([1924] 1946, p. 4) also writes that we should exclude ‘food and clothes essential for the maintenance intact of the labour force’ from national income, while Marshall ([1920], 1949, p. 63) classifies the commodities needed for the maintenance of workers as ‘consumption capital’. Nevertheless, from a social point of view, Marshall ([1920], 1949, pp. 61, 68, 434)—like later Pigou—considers the subsistence wage part of net income, and criticizes the ‘old English tradition’ of putting it into capital (Marshall, [1920] 1949, p. 649). 15 Sraffa notes that Smart ([1899] 1916) does not realize that interest is net and wage gross of ‘subsistence’, as highlighted by the possibility of negative interest, and the impossibility of a negative wage [D1/60: 2]. Sraffa also annotates the reference to Smart (1891, p. 22) and writes that ‘(t)he purpose of the sheep is—mutton, the purpose of labour is—the labourer’ (‘Lo scopo della pecora è l’arrosto, lo scopo del lavoro è il lavoratore’) [D1/60: 4]. Smart says: ‘it is possible to draw out a scientific catalogue of what things and amounts and conditions will put a sheep or bullock into the best condition for the market, just as it is possible to consider the human labourer as a force of so many foot-pounds. But the economic end of the sheep is—mutton, while the economic end of labour is –the labourer. That is to say, the “life” by which economists, who are different from butchers, must measure utility, is the life of a spiritual being from whom and towards whom all economic effort exists. To such a being, it is inconceivable that bread should have the highest use-value and diamonds none at all.’ 16 The idea that the subsistence fund must not be deducted to determine net income since otherwise the work of men would be made equal to that of animals can also be found in Ricci (1913, p. 122). On the definition of net income, Sraffa also quotes Battistella (1913) and Bowley (1922), according to whom some necessary expenses of workers (such as those of transportation or living in a highly rented locality) should instead be deducted to determine net income. 17 On the other hand, Fisher points out that the notion of ‘earned income’ is a derived concept since it is known only after the realized income and the appreciation and depreciation of capital are known. Most of the Italian fiscal school (including Borgatta, Cabiati and Einaudi) shares Fisher’s definition of income, unlike Battistella (1913) and Ricci (1913), who retain the concept of income as the amount of services that can be consumed without drying up their sources. On the debate on the definition of net income, see Wueller (1938, 1939). 18 According to this definition of wealth, it ‘includes slaves, but not freemen’ (Fisher, 1906, p. 5), although Fisher stated that human capital should also be part of wealth. 19 In this regard, Fisher (1896) states that only Cannan seems to follow the right definition of capital and criticizes that of Ricardo as wealth employed in production, that of J. S. Mill as wealth used with workers, and that of Bohm-Bawerk and Jevons as commodities used in the production process. 20 As in Marshall, this is needed to make utility and costs commensurable. 21 Fisher provides the example of ‘when a man is offered a pension to stop work. Voluntary retirement on a condition of two-third pay means that exertion was a disservice in the production of earnings equal to one third of the gross amount …. He points out that it is ‘(t)his deduction rather than that employed by the Committee of the British Association’ that ‘seems to be appropriate to obtain the taxable income of working men’ (Fisher, 1897B, p. 535, n. 2). 22 ‘solo nel caso in cui il rimandare il consumo è un sacrificio, fatto allo scopo di ricevere interesse: non quando il risparmio avverrebbe indipendentemente dall’interesse’. 23 ‘quel che resta dopo aver sottratto tutte le spese necessarie per la produzione e per reintegrare il fattore produttivo nella sua piena capacità, identica a quella precedente la produzione’ 24 ‘Se ‘netto’ riferito al salario significasse la stessa cosa, si dovrebbero sottrarre tutte le spese necessarie al mantenimento del lavoratore (spese indispensabili perché la produzione avvenga) oltre a quelle necessarie all’ammortamento, cioè alla reintegrazione del lavoratore (quota di assicurazione sulla ‘vita lavorativa’ ovvero, e meglio, spese per l’allevamento dei figli). Se si obbiettasse che allora dal profitto netto si debbano sottrarre le spese di mantenimento del capitalista, la risposta sarebbe: la vita del proprietario del ‘lavoratore’ (strumento produttivo) è necessario, in regime di salariato, perché il lavoratore possa produrre, la vita del capitalista non è necessaria perché il capitale produca. L’esempio della schiavitù serve a chiarire le idee perché la persona del proprietario del lavoratore è distinta da quella del lavoratore stesso: la vita del primo non è necessaria alla produzione, come non lo è quella del capitalista: la vita del secondo è indispensabile, come è l’esistenza del capitale.’ 25 At the end of the manuscript [D1/60], Sraffa notices that the notion of net wages may overcome the difficulty admitted by Marshall ([1920] 1949, pp. 110–11, especially 111, n. 2) of an infinite value of consumer surplus for the initial doses of income. 26 In the margin on the left side of this passage, Sraffa writes: ‘(Fisher)’ 27 ‘Vi sono diverse risposte: A) questa obbiezione implica che si concepisca il reddito naz{ionale} come un complesso di utilità totali e non di valori: bisognerebbe dunque sottrarre, per ottenere il reddito netto, tutte le disutilità derivanti al lavoratore dal suo lavoro: il risultato sarebbe identico a quello qui sostenuto, se non agli effetti delle quantità, certo logicamente.B) somme e sottrazione di utilità totali sono operazioni assurde e impossibili: le utilità totali sono spesso infinite: sottraendo da esse le disutilità, sempre finite, il risultato sarebbe sempre infinito. Ma a che scopo parlare delle variazioni di un reddito infinito sempre, e cioè invariabile.C) il rimborso delle spese di produz{ione} dà utilità non solo al lavoratore, ma anche al capitale: infatti se il capitale non venisse impiegato nella produzione, esso si consumerebbe molto rapidamente, e non vi sarebbe alcuna sorgente onde rinnovarlo. Se dunque è vero che il lavoratore, anche se riceve solo un salario di pura sussistenza (rimborso spese) ha una soddisfazione netta, (che in qualche modo va calcolata nel reddito naz{ionale}) — anche il capitalista, indipendentemente dal profitto, ricava un notevole beneficio dal fatto che, usandolo nella produzione, il suo capit{ale} viene continuamente reintegrato cioè conservato: il che sarebbe impossibile, salvo pochi casi, se esso non venisse impiegato. Insomma, indipendentemente da ogni surplus, sia il lavoratore che il capitale ricevono, producendo, la soddisfazione di esser tenuti in vita.In conclusione, per avere una definizione coerente di reddito, bisogna o comprendervi gli ammortamenti (spese) del capitale e del lavoro, o escludere gli uni e gli altri.’ 28 ‘Egli il godimento speciale di quelle spese, non lo ha come consumatore ‘soggetto del reddito’ ma come proprietario della sua propria persona, perché questa appartiene ai beni deperibili (....) Il proprietario di uno schiavo o di un cavallo gode del cibo che essi mangiano (oltreché del prodotto netto del loro lavoro) perché è una condizione necessaria perché il suo capitale continui a esistere’. 29 This may of course create conflicts between labour and capital and also shapes workers’ reactions in order to obtain a certain standard of living or keep it unchanged, although in this regard according to Sraffa, they do not act as calculating persons maximizing their utility (see for example D3/12/9: 31). 30 ‘necessario per mantenere la popolazione nello stato attuale di numero e di efficienza.’ Sraffa also specifies that those ‘who consider progress a permanent state of human society will consider an increase in wealth and population at a constant rate a stationary state: therefore, they will be able to deduct the portion that is required for this increase from net interest and net wages and calculate it with the costs’ (‘Chi consideri il progresso come uno stato permanente della società umana, considererà invece come stato stazionario quello di un aumento di ricch{ezza} e pop{olazione} a ‘rate’ costante: e potrà quindi sottrarre dall’interesse e dal salario netto la porzione necessaria a questo aumento e calcolarla fra i costi’) (D1/60: 13[9]). 31 This is also why, according to Marshall, the conventional necessaries must not be excluded from the standard of comfort, since, if the wage is not sufficient to buy them, the labourer may reduce the consumption of other necessities, thus impairing his efficiency. On Sraffa’s treatment of the necessary wage as also arising from habit and customs cf. D3/12/10: 19; D3/12/7: 42; and below, p. 12. 32 Here, there may be influence from Marx, even if no reference to him can be found in the manuscript D1/60. Note that Marshall (1898, p. 53) states that ‘free human beings are not brought up to their work on the same principles as a machine, a horse, or a slave. If they were, there would be very little difference between the distribution and the exchange side of value; for every agent of production would reap a return adequate to cover its own expenses of production with wear and tear ….’ See also Marshall ([1920], 1949, p. 418). In the manuscript D3/12/7: 105, Sraffa specifies, however, that on this page Marshall advances a ‘false opposition’ between a slave and a free labour economy, because ‘it is not true the surplus arises only in the world as it is, i.e. with free workers’. 33 As is clear in Sraffa (1926, pp. 540–41) when maintaining that the old cost theory of value is the best one available in conditions of free competition, he tried until 1927 to expel the subjective elements of utility and disutility from the demand and supply apparatus and to analyse separately the theory of distribution and the price determination of single commodities on the grounds that, although production costs depend on the prices of factors of production, small changes in the output of a commodity will not affect distribution. In what Garegnani called the Pre-Lectures (written in the summer of 1927), Sraffa showed, however, an increasing awareness of the limits of this method, realizing that costs depend on the prices of other commodities and are indirectly influenced by utility even though constant returns are assumed, since factor prices depend on the demand for goods when they are produced by different factor proportions. 34 In the Lecturers, Sraffa also notes that the classical economists regard wages ‘as a necessary means of enabling the worker to perform his work ….. In other words, the wages of the worker belong to the same class of necessities as the fuel of machines or the hay of horses’ (D2/4: 23). In other manuscripts, this ‘material necessity’ is distinguished by Sraffa from a ‘social necessity’ as one that guarantees a positive rate of profits in a capitalist economy. See, for example, D3/12/18: 11. 35 On the set of economic and social-historical circumstances that according to the classical economists and Marx may strengthen the workers’ bargaining position, thus changing their natural disadvantage in wage bargaining, cf. Levrero (2012, 2013). 36 See also Smith ([1776] 1976, I, VIII, p. 82), where he writes that ‘(i)n Great Britain the wages of labour seem, in the present time, to be evidently more than what is precisely necessary to enable the labourer to bring up a family’. 37 Needless to say, this does not imply that the subsistence wage does not allow savings to be made. In a letter to Trower (26 January 1818) Ricardo (1951–73, vol. 7, p. 248), for instance, maintains that ‘a man’s wages should, and would on a really good system, be sufficient not only to maintain himself and family when he is in full work, but also to enable him to lay up a provision in a Savings Bank for those extraordinary calls which you mention.’ 38 In this respect, Sraffa [D2/4: 37] observes that ‘in trades in which greater efforts are expected, higher wages will be paid, but this simply only to the extent that greater effort requires a greater amount of food in order to be accomplished’. See also Ricardo (1951–73, vol. 9, p. 17 and vol. 1, p. 96, n.), who specifies that the subsistence wage must ‘be sufficient to prompt [the labourer] to the necessary exertions of his powers’. 39 ‘Prima di tutto ‘sussistenza’ non ha mai voluto dire pura ‘necessità fisiologica’ (checchè ciò significhi) ma sempre anche necessità sociale e storica o abituale: cosa evidente, non foss’altro perchè con una riduzione del salario molti rinunzieranno a parte del necessario fisiologico prima di rinunziare al ‘superfluo’ come alcol, fumo, ecc. E poi quando uno ‘standard’ o livello è stato in vigore per un certo tempo esso diventa necessario—se si vuole il risultato.’ 40 In the earlier draft of the letter, Sraffa writes, ‘if the master wants to obtain the result’. Note that this process of modification will take at least the time of a generation, since only when a new generation of labourers ‘growing up in the mean time’ have been ‘habituated to such improvement of circumstances, [will] the advance price of labour come to be regarded by this new race of workmen as indispensable’ (cf. Barton, [1817], 1934, p. 22, n. 1). 41 Its new composition will be influenced by that of the surplus wage which is usually shaped by the processes of imitation and acquisition of more refined habits of consumption that led to consuming goods and services which were absent in the basket of necessaries (being a prerogative of the upper classes) or were present only in small amounts. 42 For a similar (but not analogous) interpretation, see Gehrke (2016). On the related issue that in Sraffa’s price system the wage rate can be taken as given even if it is above the subsistence level, see Levrero (2000). 43 Possibly the reference is to J. S. Mill’s theorem that savings must be exempted since otherwise they would be taxed twice. 44 It is worth remembering that in Smith and Ricardo, a tax on necessaries will not be paid by the workers but by other classes of society (cf. e.g. Smith, [1776] 1976, V, II, pp. 393–94; Ricardo, 1951–73, vol. 1, pp. 164–66). It should also be noted that in Ricardo a tax on savings probably implies a lower capital accumulation due to his acceptance of Say’s law. As outlined below, it is not so, however, when, respectively, State redistribution policies are taken into account and Say’s law is criticized on the basis of the principle of effective demand. 45 See also above, note 13, with regard to Ricci’s criticism of Einaudi’s view of exempting personal savings. 46 Of course, in poor countries, the detection of basic needs is easier. Here too, however, the subsistence wage may differ from a mere poverty line. 47 Note that the ‘minimum’ wage must be equal for married and unmarried workers, even if the latter would have (at least temporarily) a ‘surplus wage’ with respect to the former. This is so because otherwise the former would not find a job due to competition from unmarried workers (see Rathborne, 1920). The size of this ‘surplus wage’ will be greater or lower according to the female participation rate and male wage differential. 48 When the object of enquiry is subsistence wage, other elements in addition to consumption patterns should also be considered. Thus, Adam Smith in his time considered tobacco, rum, sugar and tea (and Ricardo tobacco and rum only) as luxuries for the poor, but he recognized allowances for rum and melassa as necessary for efficiency in some activities (cf. Smith, [1776] 1976, V, III, p. 477). 49 For instance, the American Social Security Administration defines the poverty lines for different kinds of families by determining the expense required for an adequate diet, and then multiplying it by 3 to take into account other kinds of necessary expenses. On poverty measurements, see also Citro and Michael (1995) and Istat (2009). The commodities included in X* may change according to the statistical information available. In Italy, expenses for housing are considered and in Canada also those for clothing, standardized data being available on the items of clothing purchased by sex, age and type of activity. The parameter h is then calculated according to the proportion of the other expenses in the total consumption of the poorest families. 50 A subjective valuation of the poverty line is offered by Van Praag et al. (1980). 51 In this case, reference is made to a certain amount of calories, squares metres and so on in order to guarantee health. 52 As in Townsend (1985), this is linked to Smith’s notion of necessities, which gives a social dimension of privation and is behind Sen’s approach (1983) and his idea that there is not necessarily a contrast between an absolute and relative approach to the standard of living (see Sen, 1984, p. 75). 53 To calculate the ‘relative’ sense of deprivation, Fuchs (1965) proposed a relative poverty line as half of the median family income. However, even an assessment of relative poverty is difficult because it can change with respect to some measures and not others. For instance, an increase may occur in the absolute number of people below the poverty line, but a fall in the normalized deficit of income with respect to the poverty line (see Sen, 1979). Measures like those put forward by Fuchs are usually calculated in the European Union (see e.g. Eurostat, 2014), together with poverty thresholds and material deprivation rates. 54 For instance, a negative income tax for maintenance does not refer to the necessary wage and is usually modulated by the size of the family and the ownership of capital assets (cf. Hildebrand, 1967; Levine and Rizvi, 2005). 55 Historically this usually happened when the supply of these goods by private capital was not able to guarantee any person with a fair rate (see Williams, 1936; Beveridge, 1942). 56 The National Minimum Wage Act of 1998 introduced such a minimum wage. On Keynesian grounds, it is also viewed as a measure of sustaining effective demand. See, for instance, Barbosa de Milo et al. (2012) on the Brazilian experience. For a neoclassical analysis of its effects, see instead Pigou (1924, pp. 561–63) and Stigler (1946), who recognized its utility only in the case of a monopsonistic labour market or the presence of a labour supply that is so abundant in relation to capital that wages would fall to zero in absence of a lower bound. For a criticism of the minimum wage leading to a fall in employment as stated by the neoclassical theory, see Card and Krueger (1995). 57 However, sometimes the workers’ net social wage is lower than that of the middle classes (cf. Shaikh and Tonak, 1987). 58 On the other hand, a universal basic income has both advantages and disadvantages for labourers, the latter explaining why trade unions often oppose it. It helps who does not work and has low income levels, and reduces gender inequalities, but it may favour wage lowering. Moreover, its financing may imply a high tax rate and the deletion of other social transfers, with the risk that its fiscal burden eventually falls on the labouring class. Finally, even if such an objection were to be overcome by tax progressivity in financing the basic income, it is not clear why it should also be given to the richest part of the population, how to face the risk of parasitism and a disincentive to work, and if it responds to the duties of citizenship and belonging to a community on which social rights should be based (see T. H. Marshall, 1950). For these reasons, Atkinson (1995) prefers to speak of a ‘participation income’, which would imply voluntary activity by who receives it. 59 Of course, the level of income to be exempted should consider the amount of necessary wage goods provided by the State. See above, pp. 15–18. © The Author(s) 2017. Published by Oxford University Press on behalf of the Cambridge Political Economy Society. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

Cambridge Journal of Economics Oxford University Press

http://www.deepdyve.com/lp/oxford-university-press/sraffa-on-taxable-income-and-its-implications-for-fiscal-policy-clQ0wVmqiy

Loading next page...

Publisher: Oxford University Press
ISSN: 0309-166X
eISSN: 1464-3545
DOI: 10.1093/cje/bex050
Publisher site: See Article on Publisher Site

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Sraffa on taxable income and its implications for fiscal policy

Sraffa on taxable income and its implications for fiscal policy

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Sraffa on taxable income and its implications for fiscal policy

Sraffa on taxable income and its implications for fiscal policy

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies