Keynesian Economics: Brief Overview

Abstract

Writing against the background of the Great Depression of the 1930s, Keynes was trying to develop a theoretical understanding of why unemployment could be persistent in a capitalist economy. The received theory at that time (which Keynes dubbed as classical but which today is usually termed as neoclassical) attributed this hysteresis in unemployment to the downward rigidity of nominal wages due to “money illusion” on the part of workers. We begin this introductory chapter with a brief overview of the economics of Keynes’ General Theory and discuss various attempts at its formalization and synthesis with the earlier neoclassical economics, embodied in the IS-LM framework. We then introduce the Phillips curve and its incorporation into Keynesian analysis. In the final section, we introduce open economy considerations into the IS-LM framework.

Appendix

The Keynesian theory of “effective demand” can be introduced in a number of ways. In this appendix, we explain the concepts of aggregate supply, aggregate demand and effective demand using an alternative approach developed by Fusfeld (1985), Darity and Young (1995) and Gillman (1999). We feel that even though the approach is somewhat unconventional, it has the advantage of establishing a link between Keynes’ Treatise on Money (Keynes 1930) and his General Theory, instead of treating the latter as completely independent of the former.

The aggregate price theory adopted by Keynes in the Treatise is essentially founded in the Marshallian micro-economic theory of the firm (Marshall 1920), but has a somewhat specialized interpretation of profits. This specialized interpretation of profits is usually attributed to Levy (1943) and Kalecki (1937) and known as the Levy–Kalecki identity. But it seems to have been known to Keynes while writing the Treatise (though he seems to have discarded it in the General Theory).

A simple exposition of the Levy–Kalecki identity is as follows (see Levy et al. 2008; Pressman 2008, etc.):

Assuming a closed economy, and no hoarding (so in the aggregate whatever is saved is invested), we have as an identity

$$ \begin{aligned} I \left( {{\text{realized}}\,{\text{investment}}} \right) & = S\left( {{\text{realized}}\;{\text{aggregate}}\;{\text{national}}\;{\text{saving}}} \right) \\ & = H_{\text{s}} \left( {{\text{household}}\;{\text{saving}}} \right) + C_{\text{S}} \left( {{\text{corporate}}\;{\text{saving}}} \right) + G_{\text{S}} \left( {{\text{government }}\;{\text{saving}}} \right) \\ \end{aligned} $$

(A1)

Further, corporate savings are simply retained profits which are corporate profits minus dividends paid out to shareholders giving us the identity

$$ C_{\text{S}} \left( {{\text{corporate}}\,{\text{saving}}} \right) = \pi \left( {{\text{corporate}}\,{\text{profits}}} \right) - D\left( {\text{dividends}} \right) $$

(A2)

Substituting (A2) in (A1) and rearranging, we get

$$ \pi \left( {{\text{corporate}}\,{\text{profits}}} \right) = I + D - \left[ {H_{\text{s}} + G_{\text{S}} } \right] $$

(A3)

(A3) is the famous Levy–Kalecki macroeconomic equilibrium condition, which states that corporate profits are equal to investment plus dividends minus non-corporate savings (i.e. savings by households and government), viz. $\left[ {H_{\text{s}} + G_{\text{S}} } \right]$.

It is to be noted that (A3) is a macroeconomic identity which may not hold at the individual firm level.

According to the Treatise, the aggregate price of output (P) is the average cost of aggregate output (AC) plus the average aggregate profit. In equation form this becomes

$$ P = {\text{AC }} + \left[ { I + D - \left( {H_{\text{s}} + G_{\text{S}} } \right)} \right]\frac{1}{Y} $$

(A4)

If we denote TR as the total revenue in the economy and TC the total cost, then (A4) yields

$$ {\text{TR}} = {\text{PY}} = \left( {\text{AC}} \right)Y + \left[ { \left( {I + D} \right) - \left( {H_{\text{s}} + G_{\text{S}} } \right)} \right] = {\text{TC}} + \left[ { I - \left( {H_{\text{s}} + G_{\text{S}} } \right)} \right] $$

(A5)

While in the short run, $\pi$ could be nonzero and $\left( {I + D} \right)\;{ \lesseqgtr }\;\left[ {H_{\text{s}} + G_{\text{S}} } \right]$, and correspondingly ${\text{TR}}\;{ \lesseqgtr }\;{\text{TC}}$, in long-run equilibrium the Treatise imposes the condition

$$ \pi = 0\;{\text{and}}\;D = 0 $$

(A6)

From (A1)–(A6), it follows that in long-run equilibrium

$$ I = S = \left( {H_{\text{s}} + G_{\text{S}} } \right) $$

(A7)

i.e. the distinction between non-corporate and aggregate saving disappears and investment is equal to both. Correspondingly, in the long run

$$ {\text{TR}} = {\text{TC}} $$

(A8)

If average costs do not change with the level of output (as would happen with a constant returns to scale production function such as the Cobb–Douglas), we may take without loss of generality this constant average cost as 1, so

$$ {\text{TC}} = Y $$

(A9)

Further, the total value of output is given by TR which is divided between consumption C, investment I (ignoring the government sector at the moment)

$$ {\text{TR}} = C + I $$

(A10)

In the General Theory, Keynes made two fundamental assumptions: (i) consumption increases as income increases but by less than income and even at zero income levels consumption is positive and (ii) investment I moves independently of Y. Putting (A9) and (A10) together, we get the famous Keynesian equilibrium condition

$$ Y = C + I $$

(A11)

Plotting (A9) and (A10) in a diagram (with these assumptions), we get the famous Keynesian cross (see Fig. A.1). The total cost curve (A9) becomes a 45° line through the origin while the total revenue curve has a positive intercept on the y-axis and then bends towards the x-axis. The point of intersection of the two curves (point K in the figure) represents the point of long-run equilibrium, with π = 0. To the left of K, the TR curve lies above the TC curve and corporate profits π are positive. Similarly, corporate profits are negative to the right of K. The output $Y^{*}$ corresponding to K is called the level of effective demand in the General Theory.

Fig. A.1

Keynesian cross

The aggregate demand price (ADP) corresponding to an employment level E is simply the total sum of proceeds expected from the sale of the output Y producible by E. If P is a suitably defined average price of all the goods produced in the economy (N(E)), by employing E amount of labour,

$$ {\text{ADP}} = PY = PN\left( E \right) $$

(A12)

In the General Theory, Keynes assumed P to be constant and capital stock characterized by excess capacity—an assumption which is not unduly restrictive as he was mainly concerned with the short run in an economy characterized by depression. However as E increases, diminishing returns may set in at some stage so that the output produced by an additional unit of employment is likely to fall. This means that the ADP curve plotted in the diagram (Fig. A.2) is upward sloping but arches towards the x-axis as E increases beyond a point. Note that we make a distinction between the AD curve and the ADP curve.

Fig. A.2

Aggregate demand price, aggregate supply price and effective demand

The aggregate supply price (ASP) corresponding to an employment level E is simply the total costs involved in producing the output corresponding to E, so that

$$ {\text{ASP}} = \left( {\text{AC}} \right)N\left( E \right) $$

(A13)

where AC is the average cost corresponding to the level of output produced by E.

As employment increases, AC is likely to rise (mainly because of rising wages and capital rentals), and coupled with diminishing returns this gives the ASP an upward slope which becomes steeper to the right, until at the full employment level $E_{\text{F}}$, the ASP becomes vertical (since it is impossible to expand output beyond this point).

The ADP curve lies above the ASP curve near the origin (i.e. profits are positive), and given the shape of the two curves, they will intersect at the point A*, and the corresponding employment level is denoted by E*. The level of output N(E*) implied by E* can be refered to as the level of effective demand. There is no automatic mechanism in the General Theory that will bring E* to coincide with $E_{\text{F}}$. In situations of overall pessimism in the economy, the ADP curve will shrink downwards and in all probability E* will lie to the left of $E_{\text{F}}$. In this case, the difference $(E_{\text{F}} - {\text{E}}^{ *} )$ corresponds to what Keynes calls as “involuntary unemployment” comprising those who are willing to work at the prevailing wage rate but unable to find employment.