I don’t often have the opportunity to publicly engage in general speculation about our economic future. I hope you don’t mind my using this event as such an opportunity. Everything I say below is offered as a mere speculation, whose establishment would depend on a great deal of interdisciplinary research. I believe, however, that posing questions and reflecting at this broad and speculative level is valuable once in a while, and I don’t know who should be responsible for doing it if not people in philosophy departments.
In my book on debt, I focussed on debt as an interpersonal relationship between a creditor and a debtor. I left largely out of consideration the intrapersonal analysis of debt, as a relationship between an earlier and a later self. A debt contract is an undertaking that imposes a cost upon a future self. Where an interest payment is required, the debtor will only agree to it if she discounts the future. If, for example, I borrow $100 today and contract to repay $110 in a year, then my willingness to make the exchange implies that $110 in a year’s time is worth the same to me as $100 now: a 10% annual discount rate. The creditor, on this analysis, is simply the intermediary of a deal between an earlier and later self. “Me at this moment and me this afternoon are indeed two”, wrote Montaigne. If you deny this, insisting that my present and future self are one and the same, then you must conclude that I make a fool’s bargain paying $110 in exchange for $100. But if you concede that present and future selves are in some way distinct, then you must concede that the present self exercises an absolute monarchy over all future selves, who are bound by debt contracts to pay whatever the present self agrees they will pay. This has an air of paradox about it, as I have argued elsewhere (Douglas 2020). But here I want to suggest that it provides a useful frame for thinking about economic institutions.
Discount rates can also be applied to the costs and benefits of possible future selves. Michael Otsuka suggests that insurance and pension schemes can be thought of as “transfers within the possible future lives of each individual — as transfers from one’s more fortunate possible future selves to one’s less fortunate possible future selves” (Otsuka forthcoming). Absent any discounting, the rational amount to pay, to insure against a 2% risk of losing a piece of property worth $2000, is $40. But I might be willing to pay more or less, depending on how I discount the avoided cost to my less fortunate possible future self. An insurance firm, pension fund, or casino is simply the intermediary of a deal between a present self and various possible future selves.
We can also extend discount rates beyond the agent, to interpersonal or social preferences. If I would give up only $1 for you to have $20, then I apply a 1900% social discount rate to your welfare. To the extent that we care about others at all, we apply some finite discount rate to their welfare, just as, in standard economic models, we care about our future and thus apply some discount to our future welfare. Economists sometimes merge the two; I once read in a textbook, for example, that the justification for giving infinite lifespans to agents in models is that real-life agents are concerned with the welfare of future generations, though discounted.
Applying the discount rates, we can model a whole economy from the point of view of one agent at one point in time. My current welfare includes everybody’s welfare, even possible future agents, all with various discount rates applied. It would of course be very arbitrary to define the welfare of an economy in terms of the preferences of a single agent at a single point in time. But is it even sensible to define the welfare of a single person in that way — taking the preferences of the agent at a particular point in time and discounting the future or possible future?
It is debatable whether time-discounting is rational. I can justify paying $110 in exchange for $100 if I say that the $110 is a future cost whereas the $100 is a present benefit, and that things in the present are better than things in the future. Spinoza argues that this is thoroughly irrational: the person guided by reason will not value things differently depending on whether they are past, present, or future (Ethics 4p62). In any case, the cost when it is paid will be a present cost. I, as a continuous agent, might contract the cost in future money, but I will pay it in present money. Joan Robinson suggests that the irrationality of time-discounting comes down to overlooking this fact. She calls it: “an irrational or weak-minded failure to value the future consumption now at what its true worth […] will turn out to be” (Robinson 2013, 394). The same debate, I think, could be had over the rationality of discounting more or less fortunate selves merely on the basis of their temporal and modal distance, rather than adjusting costs and benefits to perfect risk-weighted equality (leaving aside cases of genuine uncertainty).
The absolute monarchy exercised by the present self over future selves is absent from many interpersonal cases. It is, however, found where, e.g., a sovereign has the right to tax or fine citizens to pay for public services. More to the point here, we might also want to say, as e.g. James Galbraith proposes (Galbraith 2008, ch.12), that present generations can impose costs upon future generations who, not yet existing, have no say over the matter. But in trying to say this we will run into Derek Parfit’s “non-identity problem” (Parfit 1986, pt.4). We can avoid this by shifting to an admittedly crude way of thinking of an economic system by treating it as a giant agent. Its “decisions” to consume, invest, or risk the resources it has available are determined by its social discount rates, which can be viewed as deals that the society makes with its future self. At this level, any non-zero rate of time- or risk-discounting seems obviously irrational. Palaeolithic hunter-gatherer societies appear to have hunted their megafauna to extinction: they ate their future, as Tim Flannery evocatively put it (Flannery 1994). The future cost was significant: it forced a transition to agricultural societies in which the labour-cost of food production was initially greater (Sahlins 2017; Weisdorf 2003), to say nothing of the deaths that must have been caused by food-shortages during the transition. Supposing that hunter-gatherer societies could have avoided this fate by sustaining a lower level of overall consumption, it would be odd to argue that they did the rational thing on the grounds that the present benefits of extra consumption equalled the extreme future costs — discounted for temporal distance. If we speak of a society, like an individual, as something that exists through time, then we have to say that the society paid in present costs, not in future costs. Of course a society that ends up hunting a species to extinction doesn’t make a conscious decision to enter a contract, the way that I do if I agree to a loan at interest. All the same, we can assess the success or failure of economic systems in terms of how they can avoid discounting the future. The same goes, I think, for discounting risk — whether in an adverse or proverse direction.
Turning to capitalism, it seems possible, though I don’t mean to assert it as fact, that the prevalence of exploitation serves to keep the rate of both time- and risk-discounting down. Capitalism runs on exploitation in a sense that can be quite straightforwardly expressed, as long as we remain at a certain level of analysis. The sense is this: workers consume less output than they produce, and the surplus output is distributed to a privileged capitalist class, to be partly consumed, partly invested, and partly paid out as rent (insofar as the rentier class is distinct from the capitalist class), where again a decision is made to balance between consuming and investing. How much is invested depends in part on the discount rates of the appropriating classes, with respect to both risk and time. This notion of surplus is of profound importance for understanding the long-term social and environmental implications of capitalism. Historically, very important social transformations have been explained in terms of the capacity to generate economic surpluses of various forms (Diamond 1999; Torpey 2017).
A notorious problem arises, however, when we move the notion of surplus to a more precise level of analysis. The problem is in effect an index-number problem. We can define a surplus by saying that workers consume less output than they produce, but what do we mean by “less”? In a complex capitalist system, it is not that the workers consume only a portion of their own products and hand over a surplus to the capitalists. The case is not like that of the agricultural surplus that led, for instance, to urbanisation, the development of commercial and priestly classes, etc. This was simply a matter of the farmers producing more food than they ate, handing over the rest to the urban, ruling, and priestly classes. But under capitalism workers often do not consume what they produce at all — they might be workers in a Rolls Royce factory. To define an economic surplus here, we need to say that the value of what they consume is less than the value of what they produce. The notorious puzzle is: what is the dimension of value here? What is its unit of measure?
Marx famously attempted, like Ricardo, to use labour-time as a unit of value. Unlike Ricardo he used this specifically to define an economic surplus. But, in the first place, it is difficult to measure the labour-time embodied in different commodities. Price, by Marx’s own admission, would only be a reliable guide in a primitive system of “simple reproduction”, not in an advanced capitalist system. Moreover, because of training and education, labour is embodied in labour in differing proportions. The output of an educated worker is really labour squared: the labour of the educators enters into the value of the final worker’s labour. But, as Jon Elster points out, this means that an increase in the relative price of labour power or wage-goods with respect to other commodities will change the labour-values of commodities in different proportions, depending on the level of education of the workers that produce them (Elster 1985, §3.2.1). And this in turn means that labour cannot provide a uniform dimension of value for every commodity. Moreover, as Joan Robinson pointed out (and later G.A. Cohen, without acknowledgement), the ratio of labour-time to a given volume of output can vary, since techniques are always developing and vary from place to place (Robinson 1966, 19–20; Cohen 1979). Therefore labour-time doesn’t solve the index-number problem, which stands in the way of defining an economic surplus under capitalism.
If we think in neoclassical terms, using relative prices, value becomes a dimensionless variable, as Stanley Jevons argued (Jevons 1888, 83–84). It makes no sense to ask what the value of x is; value only expresses a ratio between goods. We cannot then give any content to the independent terms “the value of what is consumed by the worker” and “the value of what is produced by the worker”. The idea of economic surplus drops out altogether.
Alfred Marshall resurrected the notion of surplus in a slightly different way: if each worker is paid the marginal product of labour, while producing (on average) the average product, then the surplus accruing to the capitalist is the difference between the average and marginal product (Marshall 2013, Appendix K). But there are deep conceptual difficulties in defining the marginal product of labour: it is defined as the increment to output from the addition of another unit of labour, keeping capital constant. But the “Cambridge Capital Controversies” showed that it turns out to be much harder than it initially seemed to say precisely what “keeping capital constant” means (Harcourt 1972; Hausman 1980).
Relative prices, in neoclassical theory, are said to be “subjective”; they are determined by the preferences of agents rather than fixed to some “objective” standard. The relative price x/y defines how many units of y one can acquire on a market in equilibrium by supplying a unit of x. The relative price of what workers consume relative to what they produce thus defines how much output the market demands for the input. But since workers consume a heterogenous bundle of commodities, and produce a different heterogenous bundle, we still face the index-number problem in trying to compare units produced versus units consumed. On the other hand, there is a sense in which the value-ratio of input to output is unity: by supplying the inputs into the market system, the owners of the means of production are able to claim the whole output. If we could say that, for instance, the input comprises 40 units whereas the output comprises 50 units, then we would be able to calculate a surplus, e.g. 10 units. But the index-number problem deprives us of such units. Thus, although we can say that the total output has the market value of the total input, we cannot yet say how much of this output is surplus.
Some suggest that surplus can be defined as the excess of output over what is necessary for replacement of commodities. This is distinct from Marx’s concept of surplus as output whose value exceeds that of what is needed for workers to reproduce their labour power, which ran into the index-number problem. Rather, we can try to think about commodities that are not used in the production of other commodities as “surplus” — this is the line taken by Pierro Sraffa (Sraffa 1960, §6). The problem here is with the concept of “being used in production”. If a worker likes to watch sport on television to unwind between shifts, who are we to say that the television, the professional sports teams, and everything else that goes into this apparent “luxury” commodity is not being used in production? We might suppose that the worker could do just as well without it, but are we sure? Perhaps a new apparatus of coercion and surveillance would then be necessary to ensure the worker’s continued contribution, so that the elimination of the “luxury” is not in fact costless. Nor is the observation that workers got by in the past without such luxuries decisive here, unless we believe that technological and social changes leave humans entirely unchanged.
To take another example, Paul Cockshott regards the production of armaments as entirely unproductive (Cockshott 2019, §5.10.1), yet some might argue that they provide a vital good — protection from invasion — that is in fact a precondition of all other productive activity.
Others attempt to define surplus as output beyond what is needed for the simple reproduction of a society. As Michel Aglietta puts it: “To speak of reproduction is to show the processes which permit what exists to go on existing” (Aglietta 2000, 12). But how do we decide when a society “goes on existing”, and what is required for this? There is a clear sense in which colonised indigenous societies continue to exist, but there is an equally clear sense in which they are swallowed up into the society of the colonising power and transformed into something else. Some argue that immigration functions to preserve societies with low birth-rates by replenishing their source of labour; others argue that the replacement of a homegrown labour force with an imported one entails the replacement of one society by another. These debates can be very unsavoury. They are also interminable, since there are no clear rules on how we must define a society.
I propose an alternative, which is to use the notion of a social discount rate to define an economic surplus. Output appears at the end of a productive cycle; input is consumed throughout the cycle. Of course there are many overlapping productive cycles, but let us stylise them down to one, and also propose that input is consumed right at the start of the cycle. Now we can say that the input is equal in “market” value to the output, but we must apply a combination of time- and risk-discounting to the output. The society as a whole discounts the future (or various possible futures), so that its output only equals the value of the input at some rate of discount. The economic surplus is the difference between the volume of output that would equal the input without discounting and the volume of output actually produced. When the society reaches the end of the production cycle, the output that was discounted as possible future output becomes actual present output, and thus exceeds what was demanded as equivalent to the input. This allows a proportion to be wasted, stored, or productively invested.
This notion of the economic surplus as the discount on output can then be used to think about the dynamics of capitalism. By discounting the value of future output, society is motivated to invest in producing more than a mere replacement level of output. This means restraining consumption, so that the surplus thus defined is, in effect, society’s reward for the sacrifice of abstinence. This is the term Marshall used to define the return on capital, but as Maurice Dobb pointed out:
If one uses “sacrifice,” or “abstinence,” in any sense that is at all fundamental, then it is not the rich men of the world who do the “sacrifice” involved in capital accumulation. The “sacrifice” rests in the lowered incomes and narrowed consumption of the proletariat which permits the propertied class to enjoy its privileged income (Dobb 1932, ch.10).
The connection between exploitation and surplus can now be made clear. If capital accumulates in the hands of a privileged class, the discount rate of society will be reduced, for the simple reason that there are natural limits to what one can consume. Those who accumulate wealth, even if they live very large, will reach those limits and be left with a large surplus to invest. Dobb identifies sacrifice with “the lowered incomes and narrowed consumption” effectively forced upon the proletariat by their exclusion from the means of production. If they had more control over the economic process, society’s discount rate would be higher. If this discount rate exceeded what production could yield as a surplus, investment would not occur, no surplus would be generated, and the socio-cultural features that depend on such a surplus would disappear. At the limit, the society itself could be led into a transformation as extreme as the one that led from Paleolithic foraging to Neolithic agriculture.
One might say that this condition has already been reached, since the proletariat are no longer excluded from the means of production. Anyone in a modern economy can in principle access the means of production by borrowing to invest. But philosophers like Lisa Herzog and Jens van’t Klooster have traced how steeply credit markets are skewed in favour of the wealthy, who can supply collateral (Herzog 2017; van ’t Klooster 2018). This rationing of credit is another dimension of capitalist exploitation: it allows a certain class to extract rent through their disproportionate control over supply. Here too it might function to keep a social discount rate low — in this case the rate of risk-discounting. The collateral-rich, having more security, are able to take more risks with a given volume of capital, so that a society in which a wealthy minority controls the means of production will pursue more investment than a more egalitarian society.
Here are two ways in which, in line with Marx’s thinking at the most general level, the reproduction of a capitalist society depends on sustained exploitation. While abstinence is endured by the proletariat, the surplus that is its reward accrues to a privileged class, and only for this reason are the discount rates low enough to make a given level of investment “worthwhile” for the society. This condition continues only insofar as the structures of exploitation remain in place. These, however, are looking weak.
In the first place, the skewing of credit markets in favour of the collateral-rich is largely a matter of policy choice. The banking reforms of the New Deal in the United States transformed the whole financial system from being primarily an instrument of rent-extraction towards being a tool of capital development. These reforms were gradually eroded, but they could be reinstated. I and others have argued that by funding banks on the basis of their underwriting standards, rather than on their control of collateral, governments and central banks could make credit available to worthy investors rather than hoarders of wealth, thus undoing one key process of proletarianization.
It is politically contingent whether or not that happens. What appears more inexorable is a population dynamic. In Marx’s picture, workers under capitalism are paid enough to “reproduce their labour power” by maintaining themselves and giving birth to new workers:
The capital given in exchange for labour-power is converted into necessaries, by the consumption of which the muscles, nerves, bones, and brains of existing labourers are reproduced, and new labourers are begotten. Within the limits of what is strictly necessary, the individual consumption of the working class is, therefore, the reconversion of the means of subsistence given by capital in exchange for labour-power, into fresh labour-power at the disposal of capital for exploitation (Marx 1967a, ch.27).
Marx also proposed that, since capital is formed through the exploitation of the surplus generated by workers, an increase in the population of workers is needed for the increase of capital: “Accumulation of capital is, therefore, increase of the proletariat” (Marx 1967b, ch.25). Cockshott points out that in fact capitalism does not function so that workers reproduce their labour power; instead it has always depended heavily on importing labour: “Migrants from the countryside within their boundaries, or colonies without, fed the industrial growth of the great powers” (Cockshott 2019, 172). As capitalist economies run out of peripheries from which to draw in workers, tightening labour markets push up wages on the bottom end, reducing the surplus to be claimed as profit.
Marx believed that capitalism would not long tolerate this circumstance. If the price of labour increases without a corresponding increase in profit, then “the stimulus of gain is blunted”, and the rate of capital accumulation slows down. As a result:
The price of labour falls again to a level corresponding with the needs of the self-expansion of capital, whether the level be below, the same as, or above the one which was normal before the rise of wages took place (Marx 1967a, ch.25, §1).
But in our situation there is another relevant dynamic, involving urbanisation and rent. A general fall in the rate of profit means a corresponding rise in the price of assets, particularly real estate in desirable areas. There is a positive feedback mechanism here, whereby the difference in the rate of appreciation on desirable urban land versus that of peripheral land drives migration into urban centres. This drives the rate of appreciation of the urban real estate even higher and feeds back into the cycle. As a result, an increasing proportion of the surplus that was going to the capitalists goes instead to pure rentiers: holders of urban real estate and other coveted assets.
Rather than a search for profitable investment opportunities, the economy comes to be driven by an attempt to capture some of the wealth flowing to rentiers. Workers move into urban centres and seek work providing for the urban landlords. But the provisions they can offer come largely in the form of luxury services: high-end restaurants, boutique retail, five-star hotels, craft beer festivals, spas, dog groomers, various forms of therapy, etc. In other words, the survival of the proletariat comes to depend on convincing the owners of the means of production to consume at a high level. Rentier capitalism puts upward pressure on the social discount rate. Thus Marx’s “stimulus to gain” does not reappear.
Through very different in political terms, both these developments function to break down one mechanism by which capitalist society could impose abstinence on itself: namely by concentrating an economic surplus in the hands of those who have a strong appetite for investment, meaning a low discount rate, both of the future and of risk. If current technical conditions do not allow for the production of surplus matching a higher social discount rate, society faces the choice of improving these technical conditions or facing a gradual depletion of its productive capacity — a general immiseration of society. This general immiseration might be hidden for a long time by increasing inequality: if an increasingly greater share of output is claimed by those whose voices dominate the media, the arts, politics, and academia, then a general decline in overall output might not be immediately visible.
The other possibility, improving technical conditions, must in practice mean developing technologies that allow for a higher level of current consumption at a lower level of future cost. Renewable energy is a paradigm case, since consumption of wind or solar energy in principle imposes no net cost on the future at all; the death of the Sun is not brought forward by the use of photovoltaics. The use of nuclear power imposes heavy costs on a possible future, though the safer reactors (Gen III-IV) much less so. These are technologies that, for the most part, only governments can provide (Mazzucato 2018). Renewable energy provision is inextricably bound up with infrastructure and requires central planning and metering decisions, while nuclear energy cannot be left to private providers for obvious security reasons.
Keynes believed that a low rate of capital accumulation would create the opportunity for grand, long-term public investments (Keynes 1960, ch.24). Yet to achieve this, governments need access to the economic surplus. Their traditional means of access has been taxation. Taxes, however, traditionally catch flows of capital: they siphon off portions of private income. In the modern rentier economy, however, wealth-accumulation often involves no flows of income; it occurs, rather, through the appreciation of stocks: privileged assets, which tax accounting treats as savings. Psychologically, there is a big difference between the state siphoning off a flow of income and the state dipping into a stock of savings. So far governments have managed to extract the economic surplus locked into assets only by taking a portion of the flow when they are bought or sold.
Some Modern Monetary Theorists propose that governments can spend the money locked up in savings without having to tax it. They can do this by fixing interest rates and running deficits that match the rate at which the non-government sector saves. This is a clever and indirect way of spending savings de facto: creating liabilities to match the assets and occupying the “fiscal space” that is cleared by the withdrawal of income into savings. But it is a dangerous strategy, since if the private sector moves into dissaving and the public sector does not immediately cut back, the result could be a fall in the relative price of the currency, with all the distortion and confusion this might cause. It would be safer be to tax assets de jure, but that, again, involves passing a psychological hurdle that has so far not been passed.
Even if it can be passed, there is the question of whether a government will be motivated to make the investments necessary to achieve sustainability. There is a well-known literature on the alleged subjection of governments, especially elected governments, to time-inconsistency (see for example Carlin and Soskice 2014, §14.5), allegedly leading to excessive consumption. I am sceptical about much of this literature, which strikes me as excessively formal and too clever by half: underestimating the effect that cultural shifts within institutions might have. Nevertheless, if the working theory here is that egalitarian societies run high social discount rates (as the example of the Paleolithic foragers suggests) then it must follow that a society in which investment decisions are subject to democratic choice is liable to run a high social discount rate: since everyone gets one vote and all in principle have equal influence over investment decisions.
What this leaves out, however, is the effect of social transformation. Unless we subscribe to a very strict reductionism, we must admit that social conditions shape the character of the individuals that compose them. Ian Morris proposes that cultural values as well as individual moral beliefs are strongly determined by the energy source a society depends upon, and the social order it must build in order to extract it (Morris et al. 2015). We can hope that a post-capitalist, post-fossil-fuel society will foster individuals whose character and values allow for a lower social discount rate: individuals who value building for the future more and consuming in the present less. It might be that our cultural institutions need to develop in a way that transforms our own psychology: to bring us closer to Spinoza’s model of the rational and blessed agent, who values the future, the present, and the past all at the same level and, regarding herself as eternal, does not run out of concern for the future at a certain point. Those of us who spend a lot of time teaching Spinoza might not be as useless as we seem.
Aglietta, Michel. 2000. A Theory of Capitalist Regulation: The US Experience. London: Verso.
Carlin, Wendy, and David Soskice. 2014. Macroeconomics: Institutions, Instability, And The Financial System. Oxford: Oxford University Press.
Cockshott, Paul. 2019. How the World Works: The Story of Human Labor from Prehistory to the Modern Day. New York: Monthly Review Press.
Cohen, G. A. 1979. “The Labor Theory of Value and the Concept of Exploitation.” Philosophy & Public Affairs 8 (4): 338–60.
Diamond, Jared. 1999. “The Worst Mistake in the History of the Human Race.” Discover Magazine, May 1, 1999. https://www.discovermagazine.com/planet-earth/the-worst-mistake-in-the-history-of-the-human-race.
Dobb, Maurice. 1932. An Introduction to Economics. London: Gollancz.
Douglas, Alexander. 2020. “Debt and Paradox in the Early Modern Period.” In Early Modern Debts: 1550–1700, edited by Laura Kolb and George Oppitz-Trotman, 331–50. Palgrave Studies in Literature, Culture and Economics. Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-59769-6_13.
Elster, Jon. 1985. Making Sense of Marx. Cambridge: Cambridge University Press.
Flannery, Timothy. 1994. The Future Eaters: Ecological History of the Australasian Lands and People. Chatswood: Reed Books.
Galbraith, James K. 2008. The Predator State: How Conservatives Abandoned the Free Market and Why Liberals Should Too. New York: Free Press.
Harcourt, Geoffrey. 1972. Some Cambridge Controversies in the Theory of Capital. Cambridge: Cambridge University Press.
Hausman, Daniel M. 1980. Capital Profits and Prices: An Essay in the Philosophy of Economics. New York: Columbia University Press.
Herzog, Lisa. 2017. “What Could Be Wrong with a Mortgage? Private Debt Markets from a Perspective of Structural Injustice.” Journal of Political Philosophy 25 (4): 411–34. https://doi.org/10.1111/jopp.12107.
Jevons, William Stanley. 1888. The Theory of Political Economy. London: Macmillan.
Keynes, John Maynard. 1960. The General Theory of Employment Interest and Money. London: Macmillan.
Klooster, Jens van ’t. 2018. “How to Make Money: Distributive Justice, Finance, and Monetary Constitutions.” Cambridge: Cambridge University.
Marshall, Alfred. 2013. Principles of Economics. Palgrave Classics in Economics. London: Palgrave Macmillan. https://doi.org/10.1057/9781137375261.
Marx, Karl. 1967a. Capital I. New York: International Publishers.
— — — . 1967b. Capital II. New York: International Publishers.
Mazzucato, Mariana. 2018. The Entrepreneurial State: Debunking Public vs. Private Sector Myths. London: Penguin.
Morris, Ian, Margaret Atwood, Christine M. Korsgaard, Richard Seaford, and Jonathan D. Spence. 2015. Foragers, Farmers, and Fossil Fuels: How Human Values Evolve (The University Center for Human Values Series): 41. Edited by Stephen Macedo. Princeton: Princeton University Press.
Otsuka, Michael. forthcoming. How to Pool Risks Across Generations: The Case for Collective Pensions. Uehiro Series in Practical Ethics. Oxford: Oxford University Press.
Parfit, Derek. 1986. Reasons and Persons. Oxford: Clarendon Press.
Robinson, Joan. 1966. Essay on Marxian Economics. 2nd ed. London: Macmillan.
— — — . 2013. The Accumulation of Capital. London: Palgrave Macmillan.
Sahlins, Marshall. 2017. Stone Age Economics. London: Routledge.
Spinoza, Benedict. 2020. Oeuvres. Edited by Fokke Akkerman and Piet Steenbakkers. Translated by Pierre-François Moreau. Vol. 4. Paris: Presses Universitaires de France.
Sraffa, Piero. 1960. Production of Commodities by Means of Commodities: Prelude to a Critique of Economic Theory. Cambridge: Cambridge University Press.
Torpey, John. 2017. The Three Axial Ages: Moral, Material, Mental. New Brunswick: Rutgers University Press. http://www.jstor.org/stable/j.ctt1mtz5rc.
Weisdorf, Jacob L. 2003. “From Foraging to Farming: Explaining the Neolithic Revolution.” 03–41. Discussion Papers. Discussion Papers. University of Copenhagen. Department of Economics. https://ideas.repec.org/p/kud/kuiedp/0341.html.
 (Spinoza 2020, 4:416)
 One study suggests that although immigration into the UK yields economic benefits, its depressive effects on wages, as it reduces tension in the labour market, are disproportionately strong on the lower end: https://migrationobservatory.ox.ac.uk/resources/briefings/the-labour-market-effects-of-immigration/