Janus Ante Portas

1. Classical Theory of Distribution

Neoclassical or marginalist economic theory has dominated the economic literature ever since the 1870s. Making possible the simultaneous determination of prices, quantities, and all distributive variables (Panico 1988: 2), it is not only aesthetically pleasing in its universality, but it has been able to effectively “marginalize” the issue of distribution. Historically, neoclassical theory can be shown to derive from a generalization, to all factors of production, including capital, of the theory of rent in terms of land of uniform quality and intensive margins (Kurz 2003). Roughly speaking marginalist theory is simply a universal version of the Ricardian theory of rent on land, where a single principle, essentially scarcity, is said to govern the remuneration of all production factors (Quadrio-Curzio 2003). Distribution in this system essentially depends on relative factor endowments, with the rate of remuneration of a production factor being inversely related to the quantity of that production factor (Garegnani 2003). As noted in the Cambridge capital controversy, however, quantifying heterogenous capital is no mean feat, ² and the “quantity” of capital might well be found to depend on the interest rate, instead of being used for its traditional purpose, to determine the rate of interest (Sraffa 2003). One might also wonder whether income distribution in reality is really quite so simplistic, hinging essentially on one universal principle, and if this is not the case, may we not find a less universal but perhaps more robust approach?

In short, the answer is yes. Before marginalist theory established itself, another approach to economic theory had been remarkably well developed: the “surplus” approach. First formulated by the French physiocrats and later on by the English classical political economists, distribution and the theory of value both occupy a central role within this approach (Panico 1988: 4). Surplus theory, unlike its marginalist counterpart, does not determine all distributive variables, together with equilibrium prices and quantities, endogenously and simultaneously (Panico 1988: 5). Factor remuneration becomes the determining instead of the determined factor, with either the wage or the profit rate known in advance, determined outside the sphere of production (Sraffa 1963[1960]: 39). Classical authors often took the wage rate as given, with profits constituting the residue; however, within the same theoretical apparatus, one could just as easily take profits as the independent variable with wages representing the residue (Panico 1988: 5). While these variables are taken as independent by the theoretician, in practice they are far from that, with different interest groups constantly trying to shift the parameters of the system in their own favor. If we take this state of things for granted, it is handy to have an apparatus in place which can explain the consequences of these shifts if not the shifts themselves. No good can come from trying to find and define overly simplistic rules supposedly governing income distribution that would in all likelihood just cloud the issue by portraying an inherently evolutionary process in a deterministic fashion.

Sraffa (1963[1960]) has, quite rightly to my mind, posited that the profit rate might somehow be influenced by the interest rate; yet even that relationship is more elusive than is usually perceived. Typically the interest rate is seen as a distributive parameter between two groups of capitalists. In this paper I show that the rate of interest influences labor remuneration as well. As a distributive variable the interest rate can have a profound effect on the economy, as was shown by Panico et al. (2012); of course it itself is prone to exogenous determination and its value at any given point in time reflects past distributive conflicts that manifest themselves in current monetary policy rules and financial regulation, with the latter heavily influenced by lobbying activities. More will be said on the exogeneity of the interest rate later on. Similarly, the wage rate is another variable highly dependent on past bargaining position of workers and capitalists, with these bargaining positions themselves depending on a wide array of institutional and cultural factors (Garegnani 2003). As opposed to having a single determinant, this system has many determinants that manifest themselves in the form of these distributive variables.

At its very core, classical theory is open to various influences, which, ironically, grants it a sort of universality that could never be reached by neoclassical theory exactly because neoclassical theory had been striving for universality based on a single principle. Classical theory going back to Adam Smith had not only been defined by competition on the product markets, but with respect to distribution as well (Garegnani 2003). As institutions and customs change, so do the channels of influence over the distributive variables. The rise of trade unions in the 19th and 20th centuries must have had some effect on the long-run changes in the wage rate. And their demise in the second half of the 20th century was not a spontaneous event; the high interest rates during the Volcker era were not a manifestation of some invisible hand, much like the dismantling of the barriers on international capital flows and general financial deregulation were not spontaneous, but intentional and focused measures that shifted distributional parameters. Kalecki (1943) had already anticipated that something like this would happen, where industrial and rentier capitalists would eventually find a way to get the wage rate down. So whereas we can be sure about the goals of different groups, we cannot be sure what kind of levers will be used in a given historical context. Fiscal and monetary policy are bound to play a big role, which, by the way, makes monetary policy a priori non-neutral; but we can also imagine soft, cultural factors coming into play as well, like the emergence of FOX news in the United States or the general rise of tabloid journalism at the expense of investigative journalism that is able to transcend the hysteria and create a synthesis. Even movies are a reflection of the ideological zeitgeist, and while none of these soft phenomena can be measured – well one can always construe pseudo-scientific indices – that does not mean they are irrelevant or without consequence; far from it. Again, what we see with classical theory is that changes in distribution could be influenced by a plethora of different factors and that there is no reason to presuppose time invariant functional relationships between those factors. Economic reality ³ is not ergodic (Davidson 1991), as Post-Keynesians have been saying for quite some time now. This is why the classical notion of competition is so handy, because it is dynamic in nature and allows us to suppose that the different interest groups will adapt to a given situation and will have also learned from past mistakes. The goals might stay the same – bringing the wage rate down if you are an industrial capitalist, charging high interest rates for loanable money capital if you are a financial capitalist – yet how those goals are pursued, through which institutional channels, for lack of a better word, will change with time.

Having roughly sketched some of the differences between the neoclassical and the classical approach, and explaining why it is the classical approach that should be used when dealing with complex issues like distribution of income and wealth, I want to outline, briefly, the reasoning that went into building my model. Obviously I wanted it to allow for the richness of interpretations which meant that it had to be fairly simple and not too specific when it comes to possible changes in distributional parameters, which is, I believe, in line with the tradition of classical political economy. In essence the model is able to replicate the schematic from Marx (1972 [1863]): M -> M -> C -> M’ -> M’’. In a pure bank-money world money capital is lent to the firms, which then produce goods and make a surplus, but we must deduct from this surplus the part which goes back to the banks and what we end up with is M’’. Only this latter part is distributed amongst workers and industrial capitalists; the difference between M’ and M’’ goes to the banking sector, from whence it came. I wanted to capture the difference between profits of firms on the one hand, and banks on the other, where it is quite clear that surplus is created by the former and not the latter, which is not to say, however, that the latter are not essential to the workings of modern monetary economies. In fact, the post-classical school sees credit money as a vital component of the production process (Lavoie 1992: 150). What I have been able to do is take this classical system and bring it into a formal and comprehensive stock-flow model, which is something that has not been done before. It is a simple variant of stock-flow models presented by Godley and Lavoie (2012), treating distribution in the fashion mentioned above, with wages and profits on the one hand and interest rates (rents) on the other. Additionally the household sector is made up of two classes, capitalists and workers, which is another innovation with respect to existing work that has been done in this field. For all its simplicity the model does a fairly good job at explaining how the level of output is affected by changes in the distributional variables. Now, having explained the reasoning behind my decisions, let us plunge into the heart of the matter to see how the model itself is fashioned.

2. The Model

Following a common-sensical approach by Godley and Lavoie (2012) I start by first presenting the model in the form of a behavioral transaction matrix. Bear in mind the setting is one of “pure capitalism,” to borrow from Binswanger (2009), where there is no central bank, no government, and all financial needs of the firms are met by a private banking sector.

OPEN IN VIEWER

Table 1.

Behavioral matrix.

		Firms		Banks
Workers	Capitalists	Current	Capital	Current	Capital
−CWD	−CCD	+CS
		+IS	−ID
		[Y]
+(1–θ)FFD	+θFFD	−FFS
	−IFS = ΛθFFD		IFD = IFS
		−rL–1L–1		+rL–1L–1
	+rM–1M–1			−rM–1M–1
	+PBD			−PBS
			+DLD		−DLS
	−DMC				+DMS

Each of the categories above now needs to be defined as a dependent variable. This is what I propose we do next in order to get a complete system of equations that describe the bank-money world economy, made up of two classes (more on that later); vertically integrated firms a la GE, Siemens, and the big Japanese conglomerates; with a banking sector providing loans to firms. I will explain the notation together with the equations below.

C S = C W D + C C D

I S = I D

Δ L S = Δ L D

The first three equations basically tell us that scarcity is not a symptomatic problem in this economy. Supply of consumer goods (C_S), investment goods (I_S), and loans (DL_S) is able to adapt to whatever the demand might be. Equation (3) implies that the banks are very good at their job, telling apart those who are and those who are not credit-worthy (Minsky 1986: 229). In this paper we do not concern ourselves with the “fringe of unsatisfied borrowers.”

P B D = P B S

P B S = r L − 1 L − 1 − r M − 1 M − 1

I F D = I F S

I F S = Λ θ F F

where the parameters are 0 < Λ < 1 and 0 < θ < 1.

Y = C S + I S

F F = Y − r L − 1 L − 1

Equation (4) basically states that even though all the profits from banks (PB_S) go directly to financial capitalists, they cannot demand more than is supplied. In other words, no matter how influential, financial capitalists – barring government intervention or an act from God – cannot decide what they earn. In this sense financial capitalists, or rentiers, are not “masters of their fate,” as their fate is inextricably linked to that of industrial capitalists and their willingness to take new loans for further investments. This is in line with Kalecki’s (1971) view that capitalists can decide what they spend but they cannot decide what they earn.

Investment funds (IF) as defined by equations (6) and (7) tell a similar tale of one coin with two sides. In fact we could be talking of a fake duality, because there is no demand for investment funds – unless we believe capitalists, in aggregate, to be schizophrenic – without that same amount being previously supplied. The amount supplied depends on what I have called firm funds (FF) and the two parameters θ and Λ, representing respectively the power of capitalists relative to the workers and their propensity to invest.

In our economic reality there is no need for a state, so national income (Y) is simply defined as the sum of aggregate consumption and investment. With a vertical firm structure both of these two categories must in some way funnel through the firm, which means that the national income also represents the sales of the firms. These sales – or what comes to the same thing, the national income – are then distributed to three groups. The first to get their share are the banks for interest payments on past loans. What remains is the residual, which I have very unimaginatively dubbed as “firm funds.” These funds are then distributed amongst two other groups, the workers and the capitalists, with their shares being determined by, in this model at least, exogenous factors. Realistically, I suppose, only profits would make up the residual, with wages representing certain agreements from the past (the proverbial stickyness). There are also no distinctions made between firm profits and dividends, as the whole amount of FF is distributed to one party or the other. Note that like in the classical framework, we have three income categories–wages, interest payments (rents), and profits–replicating, if you will, Marx’s (1972[1863]) schematic of M -> M -> C -> M’ -> M’’ in a stock-flow setting, representing a monetary capitalist economy.

In reality, investment funds would probably never go to the capitalists’ bank accounts in the first place, but would instead stay within the firm and be re-invested. Here, however, the point was to show that it is the capitalists who are the ones that get to decide how much is being invested, and they would be in a position to do this even if the funds were not first transfered to them, but if they had remained within the firm. They are the ones that control for parameter Λ, the willingness to invest. Disregarding transaction costs, it is really quite irrelevant how the investment funds reach their final destination.

We could easily add depreciation allowances and separate dividends from profits, with both of those two categories representing outside restraints on the firms: technical and societal, respectively. For sake of simplicity, however, I have opted against doing so, instead assuming that there exists a Λ–, representing a minimum willingness to invest that corresponds to depreciation allowances, with the further assumption that the exogenous power relation parameter θ accounts for both profits and dividends ⁴ under the same category, as they would eventually end up in the same pockets anyway.

Δ L D = I D − I F D = I D − Λ θ F F

It is worth noting that (10) can also be rewritten thus:

L = L − 1 + ( I D − Λ θ F F )

C W D = ( 1 − θ ) F F

Y D W = C W D

Looking at (10), you can see the impetus for new loans (DL_D) comes from investments (I_D). Obviously most projects get funded partially by both internal funds (non-distributed profits) and loans, the former serving to reassure bankers that the latter will be repaid. There is nothing scientific per se in these assurances that represent nothing more than a part of past profits. They fall prey to the famous rule of thumb proposed by Keynes (1937); namely that we expect the current state of things to continue into the future.

The assumptions behind equations (11) and (12) are within both the Kaleckian and classical traditions. Workers spend what they earn as opposed to capitalists who earn what they spend (Kalecki 1971: 13). There is nothing to stop us from adding workers’ savings into the mix, yet I have opted against it, because we can safely assume that workers will not be able to live off their savings for long. In their case savings usually do not represent a continuous hoard; instead we can speak of saving as being merely postponed consumption. If a family is saving for their offspring to go to college, this saving has a qualitatively different role to that of a capitalist, whose savings do constitute a continuous hoard.

Y D C = ( 1 − Λ ) θ F F + r L − 1 L − 1

Y D = Y D W + Y D C

Δ M C = Y D C − C C D

r L = r L ¯

r M = r L ¯ − ε

Δ M S = Δ L S

Equations (13) and (14) revolve around disposable income (YD) with the latter being a sum of both the capitalists’ (YD_C) and workers’ (YD_W) aggregate disposable incomes. The difference between capitalists’ disposable income and consumption constitutes saving, or what comes to the same thing, the change in the stock of money (DM_C) they decide to hold. In reality, they could opt for other holdings as well, but here the only way to store wealth is in the form of money.

We can see by looking at (16) that the interest rate (r_L) is exogenous, determined by the banks themselves in absence of a central bank. The deposit rate (r_M) is indirectly set by the loan rate, the difference between the two obviously constituting bank profits. In the short term one can reasonably assume that the interest rates are completely exogenous. To borrow an analogy from Kaldor (1982), banks find themselves in a position somewhat akin to that of a constitutional monarch, being able to set the interest rates at will, yet unlikely to extend this privilege too far for fear of losing it altogether. The last equation states that any change in the supply of money (DM_S) depends on the supply of new loans (DL_S), or to put it in a different way: loans create money and not the other way around.

A brief note about the exogeneity of interest rates is in order. What do I mean when I say the interest rate is exogenous? Certainly I could quote authors of the circuit school, Post-Keynesians, and others (Lavoie 1992) but that is not the point. Sraffa (1963[1960]: 39) notes that the interest rate is determined outside the sphere of production, and I would argue that it is only with respect to the productive powers of society that the interest rate is “exogenous”; in other words it has nothing to do with the amount of “physical capital” in a society or any such nonsense. However, I would be the first to agree that sociopolitically the interest rate is endogenous. ⁵ So in the end it all boils down to methodology. The interest rate at any given point in time depends on a specific set of historical conditions which need not, and usually do not, go on indefinitely; this is also why the interest rate is often labeled as a convention. Since there is no general theory of history, there can be no general theory of the interest rate, meaning that methodologically at least it is best to treat it as exogenous. However, when discussing the issues of distribution and the trade cycle in the next chapter, the interest rate will definitely be endogenous with respect to both of those.

C C D = α 0 + α 1 Y D C + α 2 M C − 1

K T = β Y − 1

K = K − 1 + ( I D − Λ θ F F )

The capitalist consumption (CC_D) function described by equation (19) is rather simple and consists of three parts: autonomous consumption, consumption out of the current disposable income (YD_C), and consumption out of accumulated wealth (MC_–1). Equation (20) represents some aggregate target of capital (K^T) that capitalists aspire to reach in the short run. For example, an enterpreneur might want to build a new factory and that factory represents his capital target. However, his aspirations for the future depend on the results of the past, which is why K^T is a function of past sales, the latter also representing national income in this model. We do not know how the aspirations of a single agent transcend into the aggregate relation, but it is safe to assume that past results will play a large role in determining future targets, which is essentially what equation (20) does. Moving on, note the similarities between (21) and (10a). The capital stock moves in tandem with the loan volume. Talking about a stock of capital is always problematic so perhaps we could borrow from Jevons (Keynes 1964: 321), who says “not that a factory, or dock, or railway, or ship is capital, but that it represents so much capital sunk (equation (10) or (10a), if you will) in the enterprise. Accordingly, I would not say that a railway is fixed capital, but that capital is fixed in the railway.”

While the stock of loans is homogenous and simple, its mirror image is not; it is in fact a myriad of different things, created and made possible by the issue of new loans. Loans represent a simple quantitative leap whereas the growth of capital essentially produces qualitative changes to our lives; a different, more efficient and productive, technology does not mean that more “physical capital” was used; it simply means a different production method was used. Capital in our model should be understood essentially as the change in productive forces of mankind, a sort of proxy for (mainly) technological development of societies, with the volume of loans representing its abstract monetary value.

I D = γ ( K T − K − 1 ) + Λ θ F F

Δ M C = Δ M S

The last two equations are pretty self-explanatory. Investment demand (I_D) depends on the targeted capital stock and on how much profit the capitalists are willing to re-invest in any given period. Parameter γ is simply there to remind us that in reality things do not happen instantenously; targets are not achieved immediately and they change from one period to the next. Equation (23) follows from all the others and simply states that the change in the capitalist money stock (DM_C) is equal to the money supply (DM_S) from banks. However, as Moore (1997) had pointed out, capitalists – or for that matter anybody else – do not really demand deposits; they are in fact the ones who are supplying them to banks. The supply of new money originates with new loans and therefore with new investment. Equation (23) is a truism, because the money for deposits had already been supplied in form of loans.

3. Class Struggle and a Simple Trade Cycle

With the system in place, we can start looking at the implications of the results. We have a simple yet comprehensive model that represents a classical system of distribution with three types of incomes: rents, wages, and profits. Yet whereas, for example, Ricardo’s exposition was based on what was essentially a one-good economy, where corn was both the output and the input, this model represents a monetary capitalist economy, with endogenous money. I have been able to create a stock-flow model that represents classical competition on the factor markets, between two types of capital (industrial and financial) and the workers, which is something that has not been done in this fashion before.

Within the classical surplus approach one can take either the wage rate or the profit rate as an independent variable. Originally, the wage rate was assumed to be given, and due to population pressures it was further assumed that the wage rate would remain at subsistence levels where customary influences were allowed as well (Garegnani 2003). While this might have been a stylized fact back in 19th century Britain, subsistence level wages do not constitute a general law, and there is no reason why workers could not accrue a share of the surplus large enough for the wage rate to go above that level. Within the Sraffian system (1963[1960]) there are no assumptions made about the level of wages, the surplus can be distributed in any way, ranging between the two extremes where all of it accrues to either workers in the form of wages (or profit sharing schemes of some sort) or to capitalists in the form of profits. In my model the distribution of firm funds, the part of the national income left after the repayment of interest on past loans, follows the same logic; namely that at least theoretically there is nothing to stop theta (the “power” parameter, which reflects the current bargaining position between capitalists and workers) from being either 0 – with all of the firm funds going to workers – or 1, where the opposite situation occurs. Realistically, theta will most likely vary between certain upper and lower bounds, never reaching the extremes and moving slowly in time, representing the “grinding” scramble for income as one of the defining characteristics of capitalism.

The other distributive variable, one which defines the amount of firm funds which are distributed amongst capitalists and workers, is the interest rate. Much the same way as theta could be said to represent, in the spirit of classical factor market competition (Garegnani 2003), both the current and past bargaining positions between industrial capitalists and workers, the same could be said for the interest rate, itself dependent on a myriad of different considerations, all of which are essentially aimed at maintaining or increasing the rentier capitalist’s share of the total output and at perpetuating the system that grants this class the leisurely route of capital accumulation. The interest rate primarily reflects the seemingly antagonistic relationship between financial and industrial capitalists, where it should be noted, however, that both of these groups receive the surplus, and while issues of division will, almost by definition, be divisive, it is in both their interests that the surplus is as large as possible. Marx (1972[1863]) speaks of the relationship as representing a quantitative division of gross profits between the two groups. In my model the situation could be said to be similar, except that there is no explicit “surplus” category and that there is no equivalent category to gross profits in the sense that Marx envisioned them. Net profits are only determined after we know theta and they are contested as a residual, with interest payments having already been deducted from the national income. Regardless, I will try to argue later on that Kalecki’s (1943) idea of a grand coalition between industrial and financial capitalists makes intuitive sense, given the results of the model.

Solving the model can be done without having to make use of any artificial processing power, yet nevertheless the process itself has been left for the appendix. The category – perhaps a more appropriate term than a variable – which will prove to be of interest for us in the further analysis in this paper, revolves around national income variations with respect to parameter shifts. This is why not every category from the behavioral matrix has been expressed in parameter form and why some have been, again, left for the appendix. What distinguishes this model, among other innovations, from the similar ones by Godley and Lavoie (2012) is that the interest rate parameter for loanable money capital comes directly into the results. The interest rate is an important distributive variable, one over which the financial system has direct control and one which influences the level of output.

Y * = α 0 ( 1 − α 1 ) ( ( 1 − Λ ) θ ( 1 − r L ¯ β ) + r L ¯ β ) − α 2 β

The equilibrium level of national income is captured by relation (30). At this point it is perhaps worth noting that the paper will not deal with disequilibrium dynamics, which is not to say that out-of-equilibrium positions are not worth analyzing in general, but that changes distribution, the leitmotif of this paper, which leads to changes in the level of output constitute various, different equilibrium positions of the model, even if they are not optimal from the viewpoint of employment, for example. Having a concept of equilibrium based on full employment of resources is hardly helpful, knowing that capitalism has never experienced long periods of full employment. The question of stability and the possibility of the system reaching equilibrium in the first place is dealt with in the appendix.

Now we turn to the issues concerning distribution and the trade cycle. Obviously there are different reasons for the existance of trade cycles, some of which have accompanying theories and some of which do not. The model in this paper was built to show how output reacts to changes in distribution and we can use this feature to give a meaningful interpretation of a trade cycle. It is quite clear from (30) that distributional parameters ( r L ¯ and θ), among others, influence the level of output; what I will argue is that these changes in distribution can be shown to induce a trade cycle. Obviously causality runs both ways and changes in the level of output can themselves induce changes in distribution. All we need to do is imagine successive periods of falling output, eventually leading to a higher unemployment rate, which brings down the bargaining power of workers vis-a-vis the capitalists in form of a growing theta. Unemployment is far from the only factor which could influence theta – in fact we could imagine a plethora of factors influencing the “power parameter” between workers and industrial capitalists – yet “the sack” remains a disciplinary mechanism par excellence in capitalism. It is precisely because of its role that capitalists will never be keen on protracted periods of full employment, something noted by Kalecki (1943). In terms of distribution, therefore, crises, by dint of inducing unemployment on the economic system, play a vital role in bringing the bargaining power of workers from the previous periods back down again. While this phase of the trade cycle is unpleasant and far from Pareto optimal, I would not characterize it as out-of-the equilibrium, and indeed within my framework it remains fully within it; all that changes in the model is the theta, which, should unemployment persist, will likely shift in favor of capitalists (that is, it will go up).

Theta, θ, and the interest rate, r L ¯ , have a rather simple and straightforward relationship with the national income; should any of the two variables increase in value, the level of output goes down ceteris paribus. As far as potential levels of output are concerned, a low theta, coupled with low interest rates on loanable money capital, would constitute the most beneficial social arrangement. Financial capitalists especially will not be keen on such an arrangement, where a relatively small part of national income ends up in their pockets, an unhappy consequence of low interest rates for a class which lives by those very same interest rates. Additionally high levels of output, coupled with low unemployment, eventually start shifting theta downwards, eroding the share of capitalist profits even further. While in absolute terms the whole society might be better off, in relative (and absolute) terms this arrangement favors the workers, who are able to accrue for themselves an ever-increasing chunk of the national income. I suppose the fear could be that eventually, in the long run, the capitalist class, or at least its rentier aspect, would cease to exist, something which Keynes (1997) had actually hoped would happen with a policy of continuous low levels of interest. So whereas Kalecki (1943) was, only seemingly of course, befuddled as to why capitalists were not keen on policies of full employment (even though those policies would bring about a higher mass of profits), in reality no paradox exists. Protracted periods of near full employment bring about a change in the bargaining position of workers (theta goes down), meaning a larger part of firm funds will go into the pockets of workers, which, as we will see in the next section, is crucial. The biggest fear, I suppose, is that in the long run these sort of progressive policies could destabilize the existing class structure altogether. The exchange ratio between a unit of money (Keynesian term) and a unit of labor power (Marx’s term) would change, leading to a devalution of the unit of money in terms of labor power (Patnaik 2009: 136). Essentially this means that the money wage (and the real wage) would go up (Patnaik 2009: 165) at the expense of real profits and at the expense of the existing money stock, which would devalue in terms of labor power as well. Money in capitalism has a peculiar feature, namely it starts the production circuit and it therefore has to have command over labor power to have any value for the capitalist (Graziani 1998). Furthermore Graziani (1998) notes: “The only exchange relation that involves money therefore is that between money and labor, and the only price that is formed is the wage rate. The problem of the value of money is therefore reduced to the problem of determining wages.” Therefore if the wage rate is allowed to go unchecked, it will erode the value of money (for the capitalists). However, as long as the power to invest is safely in the hands of capitalists, this is not a problem because an “investment strike” creates unemployment and eventually money regains its value compared to labor power, which is when investment will (eventually) start picking up the pace again. Now if a government were to interfere with this “natural” order of things by trying to alleviate unemployment through public investment programs, the well tried principles of capitalist business cycles would not be able to do what they were supposed to, which is to bring down the wage rate.

Fortunately there is another route, albeit a circuitous one, that can bring profits up without directly having to shift theta: the interest rate. Remember what the workers and capitalists contest are the firm funds (36), and their size varies inversely with the interest rate; the higher the rate, the lower the funds and the greater share of national income accrues to capitalists without there being any positive change in the value of theta. It should be noted, however, that there is antagonism between industrial and financial capitalists, but only to a certain point. Furthermore I would like to point out that we will not touch on all of the aspects that the interest rate has on distribution, output, and employment; high interest rates can keep down investments and employment ex ante, so to speak, meaning that since fewer projects are profitable, there will be more people unemployed and so on. This does not undermine the basic findings of the following paragraphs; I only mention it to remind the reader that not all “transmission mechanisms,” for lack of a better term, will be considered here.

F F * = α 0 ( 1 − r L ¯ β ) ( 1 − α 1 ) ( ( 1 − Λ ) θ ( 1 − r L ¯ β ) + r L ¯ β ) − α 2 β

Much like firm funds, the level of output is also inversely related to the level of interest, which means that not only will higher interest rates translate into a smaller “pot” for both workers and industrial capitalists; additionally, with a lower equilibrium level of output, the bargaining power of workers will (eventually) go down further, meaning theta will go up, so that even out of the already smaller firm funds an even lesser part will accrue to the workers. High interest rates therefore do two things: they increase the share of income that capitalists get as opposed to a situation of a low theta and low interest rates, and by making the workers worse off (lower wages, unemployment, etc.) their wage claims fall, eventually making investment more profitable again. It is beggar thy neighbor, but in a class setting. With this policy, capitalists will have seemingly empoverished themselves, only to empoverish workers even more, plunging them into desperation, thereby effectively enriching themselves. Again, what has to be noted is that high interest rates are not beneficial to all capitalists, but by putting pressure on wage claims – indirectly of course – they eventually benefit the class as a whole. We should remember also that inasmuch as every capitalist will want to acquire at least some part of their wealth in moneyed form (a constant hoard if you will), then every capitalist is at least partially a rentier and will not want to see the value of their accrued wealth erode due to ever higher wage claims. Of course industrial capitalists would ideally prefer high theta and low interest rates to prevail at all times. However, without going into too much detail, we should remember that in a competitive capitalist economy, capitalists will wish to produce as much as possible, which means that more people will need to be employed and output will rise. This process carries with it the seeds of its own destruction, because theta will eventually start falling. Therefore if high interest rates stop this process, by making old investments unprofitable and by discouraging new investments, the policy will have done its job in making sure that the workers’ share (and pari passu their wage claims) is kept down.

Essentially what this model shows is that the interest rate can influence the profit share and that the share of profits moves in the same direction as the interest rate. Sraffa (1963) mentions that the interest rate could influence the rate of profit and in Panico’s (1988) model we see exactly that: high interest rates bringing about high profit rates and vice versa. What is the profit rate in our model? Quite simply, following Graziani (2009[2003]: 103) it is the difference between output and the monetary costs of production, divided by these same costs, yielding the following result:

r = θ ( 1 − r L ¯ β )

In my model the interest rate will obviously have negative direct effects on the profit rate, indirectly; however, through its negative effects on the bargaining power of the working class (lowering output and firm funds), it will eventually have an overall positive effect on the rate of profit because of its influence on theta. ⁶ A further interesting feature of the profit rate (and the profit share as well, for that matter) is that if theta is zero, which would be optimal for society (maximum national income), the profit rate is also zero, a scenario that capitalists will do anything to avoid. In Panico’s (1988) model the positive relationship between the interest rate and the rate of profit is more clear-cut, whereas in my model the same conclusions only hold if we posit that a higher interest rate will, by lowering output, eventually influence theta. It is also worth noting that once this has been done, the interest rates can of course go down again. In fact if prospects of profitability pick up and the confidence in the system is restored, this might very well happen automatically.

However, it is not only distributional parameters that influence the level of output, or the distributional dynamics for that matter. Capitalist’s propensity to consume, α₁, also plays a major role and the higher α₁, the higher is the level of output. Obviously this parameter, together with the propensity to invest, Λ, is very closely linked to Keynesian concepts of uncertainty and animal spirits, bringing into play expectations and their influence on the level of output as well as the issue of distribution (with the two being interconnected anyway). Now were we to imagine that for whatever reason, maybe a fall in the level of output or due to some more or less rational fears about what the future might hold, the capitalists would start hoarding their income, the level of output would fall even further. So what was already a crisis might develop into an even worse crisis, something that has been pointed out lately by Keynesians of all varieties, with respect to European austerity measures. However this paradox is only paradoxical with respect to the level of national income, whereas there is nothing paradoxical about its effects on distribution; for it is very likely that a fall in national income will eventually lower the bargaining power of the workers and the capitalists’s self-imposed “austerity measures” will have done their job. Note also the effects of this so-called austerity (a negative change in α₁) on the profit rate; as the bargaining power of labor falls, which it eventually must, both the share and the rate of profit will go up. Thrift can therefore be beneficial to the profit rate, not due to some moral reasons but because we can realistically expect that a campaign of thrift will boost theta. In conclusion, capitalists can, through saving (or spending), determine the level of activity in an economy, and by doing so they implicitly influence the distribution of income as well. It should be noted, however, that a previous change in the latter could have been at the root of the capitalist’s decision to change their spending habits in the first place. Distribution, therefore, both determines and is determined by the level of output, which is itself dependent on capitalist expenditures.

4. Scarcity, Non-produced Means of Production, “Luxury” Commodities, and Distribution in Capitalism

The disconnect between a high theta (income share of capitalists), high interest rates, and low consumption on the one hand, and maximum level of output on the other, was made quite clear in the previous section. Paradoxically, even net profits of industrial capitalists, θFF^*, suffer with a higher theta. So if by increasing their share they are actually undermining themselves in absolute terms, with a lower theta granting higher profits, then why would they ever be interested in increasing their income share in the first place? While we could simply refer to the profit rate and say that no paradox exists, I believe there is another, complementary, story to be told. From a purely mechanistic point of view, the system would be best served if no income at all accrued to industrial capitalists and if the interest rates were only barely positive. There are various reasons for this state of affairs; some might have to do with human nature, some with the way we are brought up, but at least some of it is connected with the fact that there exists a whole range of commodities which are not producible. Another would have to do with control over the means of production, produced and non-produced alike. Finally, with control over the means of production one also dictates the levels and the composition of output, and, perhaps most importantly, its distribution as well.

A production circuit in capitalism starts and ends with money as was noted, among others, by Graziani (1998); money needs to be able to acquire labor where the wage rate is the price of labor in terms of money. Two things need to be mentioned: first, continuous full employment is likely to raise the wage rate, meaning the value of money will depreciate with respect to labor, which also means that the fruits of past accumulation get devalued. ⁷ Second, prolonged periods of low theta, to use my terminology, could endanger the existing social arrangement, which is why capitalists will always oppose any full employment policies. It is not in the interest of the long-term survival of the system for theta to be that low; hence, a system that produces more than any other in history could actually – again in very straightforward mechanistic terms – produce a lot more. Strictly speaking, Kalecki (1943) was correct in saying that policies of full employment, in our model typified by a low theta, would increase the mass of industrial profits, θFF^*. The problem is that this would increase all the other income categories in absolute terms as well, and what Kalecki did not take into account is that not all commodities are producible. All other things being the same, if the income shares change, so does the command over these commodities (imagine beachfront property, to give but one example). Furthermore, and perhaps more worryingly, one could even imagine workers, or groups of workers, contesting control over means of production, as was the case in Sweden with wage-earner funds (Whyman 2008).

Let us, therefore, turn our attention to the means of production. It is essential, from the capitalist’s point of view, that workers’ purchasing power be constantly kept low enough so that they are never able to contest the control over means of production and the income (essentially rents) generated by them. Control over means of production grants one indirect control over the level of employment and, consequently, over the money value of labor power; as long as labor’s share is kept low enough to not contest control of the means of production, the status quo can persist. Note that in the current scheme capitalists pay workers for wage goods – the proceeds of which end up flowing back to the capitalist class anyway – getting investment and (luxurious) consumption goods in return. If relative ratios change, the workers can afford to buy goods previously unavailable to them, meaning the capitalists lose in real terms.

Because of their scarcity, special consideration and importance should be given to non-produced means of production. With produced means of production and produced goods in general there is always the possibility that production will be able to keep up with demand and scarcity will not be the issue. The same cannot be said for non-produced goods, such as land, oil, and natural gas. By keeping the income share of workers low, one essentially makes sure that these non-produced means will never be contested by them, and the owners of these means of production will be able to extract their rents. From this point of view, high interest rates and high theta make perfect sense; capitalists can make use of both levers to make sure that their control over scarce resources is kept intact, the price being that some firms go under and that fewer production circuits get started in the first place. Again we come to the same problem, that should the wage rate go up, the playing field would be equalized. This is in the interest of both financial and industrial capitalists, which is not to say, however, that both of these groups would not compete amongst each other for control over these non-produced means of production.

My model does not deal directly with scarcity; it merely shows the supposed paradox of sub-optimal positions, ⁸ at least some of which are not paradoxical when we become aware that scarcity has not, and likely never will, go away. Ricardo was well aware of this duality; in his theory we find both scarcity connected with rent, and productiveness connected with wages and profits (Quadrio-Curzio 2003). However, from Wicksteed onwards, neoclassical theory had basically universalized Ricardo’s theory of rent to all factors. Policywise this spontaneous neoclassical determinism implies the existance of only one set of optimal policies, thereby successfully subduing any policy alternatives (Branco 2012). So part of the reaction, led mainly by the Post-Keynesians, was to deny the influence of scarcity in modern capitalist societies (Quadrio-Curzio 2003). Even if we disregard human nature, which in its competitive drive leads us to zero sum games that manifest themselves through conspicuous consumption and the like, there is no denying the fact that certain means of production are, at any given point in time, limited in quantity and, given the technological constraints, non-reproducible. While it is true that things change and technological constraints disappear, scarcity remains; oil has replaced coal, to name but one example, which just means that one scarce commodity has driven out another scarce commodity. Knowing that non-producible means of production influence prices and distribution, sometimes in unexpected ways as proven quite decisively in a Sraffian framework by Quadrio-Curzio (2003), is just another thing to consider with respect to the role of theta and the interest rate. Perhaps in order for existing social relations to continue, the system cannot operate at the maximum theoretical value of Y^*. Keeping a high theta, coupled with occasional Volcker-like interest rate shocks, goes a long way in making sure that a large part of society will never have the purchasing power and the political power to gain hold over non-produced and produced means of production alike. If the distribution changes in favor of the workers, for the same amount of scarce non-producible commodities, they can buy more of these commodities, irrespective of the fact that profits might have risen in absolute terms; if the profit share falls, that is enough.

With respect to surplus product and luxury products Sraffa (1963 [1960]) writes: “One effect of the emergence of a surplus must be noticed. Previously, all commodities ranked equally, each of them being found both among the products and among the means of production….But now there is room for a new class of “luxury” products which are not used, whether as instruments of production or as articles of subsistence, in the production of others.” Characterized as passive by Sraffa (1963[1960]), with respect to the profit rate and the price-relation between them and all other goods, this is only true within the scope of Sraffa’s own system. Sociologically we have to be aware of the importance of these products and we cannot even begin to describe, in one paragraph, what their importance to the system might be. The point is, if the system produces such goods, and there is nothing in our model saying that firms would not, among other things, produce luxury commodities as well, it is safe to assume that one would wish to be in a financial position to acquire them. So ideally you want to have a system where a Ferrari factory exists in the first place, and you want to be able to buy that Ferrari without having to manufacture it yourself. A precondition for the existance of the Ferrari factory is a large enough pool of free labor. I am not saying that no luxury goods are consumed by the working class, far from it; in general, the best of these goods will be consumed by capitalists, be they of the active or leisurely persuasion. Again, a relatively high theta gives you both; it makes sure there are workers who are willing to build your luxury sports car, and you also make sure that by building this magnificent vehicle they will not make enough money to buy it themselves. As long as the workers spend their earnings mainly on wage goods, all the other goods in the economy, scarce and abundant alike, will remain out of their reach.

These are just some of the reasons why it is highly unlikely that, given the current social system, we would see the “optimal” possible levels of output. The model makes explicit the intuition of Kalecki (1943) and the paper tries to go beyond that intuition, providing some explanations as to why certain equlibria might not prove tenable in capitalism. I do not believe that, in the long run, the coexistence-antagonism between rising wealth on the one hand, and massive poverty on the other, can be bridged, because the latter is the pre-requisite for the former; this becomes especially obvious knowing that theta has a positive relationship with the rate of profit and a negative relationship with output. That being said, the period after the Second World War was a rather successful example of how things can be done differently and why we should not give into the “tyranny of now.” On the other hand it is also true that in spite (or because) of a great crisis of capitalism, the system seems stronger than ever. The bargaining power of First World workers is falling, and, judging by the financial results of the world’s most prominent financial houses, the rentier interests seem to have recovered; everything seems to be in motion for an eventual new “recovery.” In the meantime, keeping the theta in favor of the capitalists, a global pauperized underclass has emerged; a class of price-takers, a reserve army of labor of unimaginable proportions (Patnaik 2009: xvi), that, if nothing else, constitute a constant “threat” to the employed, and as we have seen in Greece can have devastating political consequences. Remember that full employment in and of itself is not problematic; it is the improved bargaining position which inevitably follows full employment that makes these policies undesirable. So were we to one day wake up to a golden dawn of a brave new world, with a regime that would ensure full employment due to popular demand of this pauperized underclass, it is very likely that real wages would be kept down by other means. And this would not be unheard of; after all, one of the important functions of fascism was to remove capitalist objections to full employment (Kalecki 1943) by doing exactly that: keeping the appetites of workers (theta) in check. If such a system would be stable in the long run is hard to say. Due to the nature of investments undertaken in that social climate, war seems a likely scenario: destroying wealth, only to create it anew.

5. Conclusion

This paper presented a stock-flow model in the tradition of Godley and Lavoie (2012), with some innovations such as the explicit class division of households between capitalists and workers. However, the most important change lies in the treatment of distribution between different classes and amongst the two groups of capitalists: industrialists and financiers. National income in any period is divided between interest payments on past loans and the residual, called firm funds, FF, which are contested by industrial, firm-owning capitalists and workers. Thus what I have managed to replicate, in a different setting, is the simple division of the surplus, as envisioned by Marx (1972[1863]) and represented by this simple schematic: M -> M -> C -> M’ -> M’’. The comparative static analysis of equilibrium positions gives interesting insights into the role of distribution and fluctuations of output. I feel one of the main strengths of the model is that a very wide variety of influences are allowed, which can shift especially theta, θ, the power parameter determining the division of firm funds, FF, between industrial capitalists and workers. Unlike the neoclassical system, which essentially hinges on scarcity to determine factor reward, my model follows the original classical tradition, where competition is seen to operate not only on product but on factor markets as well (Garegnani 2003), with each group vying for a greater share of the spoils.

Comparing various equilibrium positions, by shifting parameters such as theta, θ, or the interest rate, r L ¯ , or even the capitalists’ propensity to consume, α₁, can be seen as a portrayal of a distributionally induced trade cycle. While uncertainty and expectational mismatches might certainly play a big role in various crises of capitalism, I also believe there is a systemic, structural, even deterministic explanation of output fluctuations which essentially hinges on issues of distribution. Some of the results reaffirm what we already know, whereas some reveal additional interesting insights. The level of output is inversely related to changes of theta and the interest rate, whereas more capitalists’ consumption boosts the level of output. So from the point of view of optimal, which is to say maximum, level of output, we have to ask ourselves, why would the capitalist ever want a higher share of the national income, if this means that, in absolute terms, everybody, themselves included, are worse off? There are various reasons, some of which were already mentioned by Kalecki (1943) in his essay Political Aspects of Full Employment; in a nutshell any prolonged period of falling theta (falling “power” and income share of industrial capitalists) and low interest rates (falling income share of rentiers) could lead to a change in the system itself. The “sack” is only a good disciplinary measure in capitalism if a certain amount of people are always unemployed. Additionally, there is the fact that the profit rate is a positive function of theta, whereas exactly the opposite is true for output, making the disconnect between what is good for society and what is good for the capitalists painfully obvious. Nobody knows exactly how a prolonged period of full employment would change the system; it is probably safe to assume that theta would be low, and with proper institutional changes the interest rates would most likely also be low. And while total income would be higher for everybody, the question is whether or not the relation between the money unit and the unit of labor would change. In other words the danger is that the accumulated wealth of the capitalist class would lose its value vis-a-vis labor. Not only that; workers could contest the control over means of production (produced and non-produced alike), luxury goods, and scarce goods. Kalecki (1943) himself had noticed that any long-term progressive social policies a la full employment would likely find themselves facing the phalanx of industrial and rentier interests alike. Obviously both the rentiers and the industrialists would prefer higher levels of output, but not at the risk of falling influence and control over labor power and its fruits. What good are higher nominal profits if one cannot hire a butler, own a beach house in the Hamptons, and enrage the ghost of Veblen with one’s conspicuous consumption?

Distributional struggle can also explain such phenomena as financialization and financial deregulation. When capitalists are faced with a given and unmovable theta, what remains is the use of the interest rate. While industrial capitalists might be adversely affected by this measure in the short run due to both a smaller share and a lower profit rate, and indeed many might actually go under due to this change in monetary policy, in the long run this course of action brings back control of the economy to the capitalist class. After all, if the division of firm funds is in favor of workers, the logical thing to do is to make these firm funds smaller, thereby negating the success of the struggle against industrial capitalists. A rising interest rate lowers the share of income going to the workers, making the real wages go down or remain constant, with no further gains being made possible. A similar result, but in a Sraffian setting, was reached by Panico (1988). Both the share and rate of profit move with the interest rate. This need not happen instantaneously since it might take a while for the bargaining power of workers to fall, raising both theta and the profit rate. This obviously means that with prices staying the same workers’ wages go down, or unemployment goes up; either way, the power relations are restored and wage claims fall. We see that industrial capitalists and rentiers are therefore merely two sides of the same coin; they are like Janus, the two-faced god of beginings, endings, and transitions: a swinging pendulum that dances to (and creates) the rhythms of the zeitgeist. As the two souls of capitalists duel with themselves, they are both the orchestra and the dancers in the great symphony of capitalism.