Mr. Keynes and the "Classics"; A Suggested Interpretation

Hicks observes that Keynes’s General Theory left “the ordinary economist quite bewildered” because it is unclear how much of it is a genuinely new theory versus a restatement of positions economists already held (p. 147). Rather than debate Keynes’s rhetoric, Hicks proposes to “construct a typical ‘classical’ theory, built on an earlier and cruder model than Professor Pigou’s,” so that, by comparing it point for point with Keynes’s own theory set out in similar form, the real points of difference – and thus Keynes’s actual innovations – can be isolated (pp. 147-148). The paper is explicitly a paper of comparison, not persuasion: “we shall at last have a satisfactory basis of comparison” (p. 148).

Working with a fixed stock of equipment, homogeneous labour, and a given money wage w, Hicks lets x, y be outputs and N_x, N_y employment in investment- and consumption-goods trades respectively, with total income I = wx(dN_x/dx) + wy(dN_y/dy) (p. 148). Assuming the “Cambridge Quantity equation” M = kI (money demand proportional to income), together with investment depending on the interest rate, I_x = C(i) (the marginal-efficiency-of-capital schedule), and saving depending on both income and the interest rate, I_x = S(i, I), gives three equations in three unknowns – I, I_x, and i – from which total employment follows (p. 149).

Because I = M/k, total income is “completely determined” by the quantity of money once k is given, but total employment is not determined by income alone, since it also depends on how income is divided between the investment and consumption trades (p. 149). An increase in the inducement to invest raises the interest rate, which raises saving to match, shifting labour toward the investment trades; total employment rises or falls depending on whether the elasticity of supply is greater in investment or consumption goods. A rise in the rate of money wages, holding k independent of wages, “will necessarily diminish employment and raise real wages” in this classical world, since an unchanged money income cannot buy the same quantity of goods at a higher price level without a fall in employment (p. 150).

Hicks notes that classical theory can only explain the large swings in money income seen over a trade cycle through variations in M (bank credit expansion or contraction) or in k, or through changes in income distribution (p. 150). Pressed further, explaining variations in k requires a theory of what determines the demand for money, and “on grounds of pure value theory, it is evident that the direct sacrifice made by a person who holds a stock of money is a sacrifice of interest,” so that “the demand for money depends upon the rate of interest” – at which point, Hicks writes, “the stage is set for Mr. Keynes” (p. 151).

Hicks writes Keynes’s system as M = L(i), I_x = C(i), I_x = S(I) (p. 152), which differs from the classical system in two ways: money demand depends on the interest rate rather than income (liquidity preference), and saving depends only on income, not the interest rate (the multiplier equation). Hicks judges the second change “a mere simplification” that is “ultimately insignificant,” but states plainly that “it is the liquidity preference doctrine which is vital” – because it makes the interest rate, not income, the variable determined by the quantity of money, with income then set by the multiplier acting on investment (p. 152).

Because “the transactions motive must always come in as well” as the speculative motive, Hicks writes the full General Theory system as M = L(I, i), I_x = C(i), I_x = S(I) (p. 153). From this he draws two curves in income-interest-rate space: the LL curve, sloping upward, showing combinations of income and interest consistent with money-market equilibrium for a given money stock; and the IS curve, derived from the investment and saving equations together, showing combinations of income and interest consistent with saving equaling investment. Income and the interest rate “are now determined together at P,” the intersection of IS and LL (p. 153) – the construction that became known as the IS-LM model. Hicks likens this simultaneous determination to the marginalist revolution’s simultaneous determination of price and quantity, replacing the older, one-directional quantity-theory logic (p. 154).

Hicks paraphrases Keynes’s proof (his own version, distinct from Keynes’s) that if the cost of holding money is negligible, “it will always be profitable to hold money rather than lend it out, if the rate of interest is not greater than zero,” so the rate must always be positive and, in the extreme, the short rate can approach zero while the long rate must lie above it – because the long rate must compensate the lender for the risk that the short rate rises before the loan matures, exposing them to a capital loss (pp. 154-155). This risk, which “provides Mr. Keynes’ ‘speculative motive,’” ensures the rate on loans of indefinite duration “cannot fall very near zero” (p. 155, incl. fn. 7). This minimum applies to any LL curve, however much the money supply is increased – an increase in M shifts LL to the right, but “the horizontal parts of the curve are almost the same” (p. 155), which is why Hicks calls the demonstration central: it explains why monetary expansion can be powerless in a slump.

If the IS curve lies well to the right (strong inducement to invest or a strong propensity to consume), P lies on the upward-sloping segment of LL, and “the classical theory will be a good approximation” – a rise in the inducement to invest raises the interest rate as well as (modestly) income and employment. But “if the point P lies to the left of the LL curve” – on its near-horizontal, liquidity-trap segment – “then the special form of Mr. Keynes’ theory becomes valid. A rise in the schedule of the marginal efficiency of capital only increases employment, and does not raise the rate of interest at all. We are completely out of touch with the classical world” (p. 154). Hicks concludes from this that “the General Theory of Employment is the Economics of Depression” (p. 155) – it is the special, liquidity-trap-adjacent case, not the fully general system, that captures what is distinctive about Keynes’s policy conclusions.

Hicks observes that Keynes’s own exposition simplifies by measuring things in “wage-units” so that only changes in money wages, not other changes in income, shift the investment schedule; Hicks questions this and reinstates income as an argument of investment too, writing the fully symmetric system M = L(I, i), I_x = C(I, i), I_x = S(I, i) (p. 156). To derive the resulting more general IS curve, Hicks constructs, at a given level of income, a CC curve (the marginal efficiency of capital in money terms at that income) and an SS curve (the supply of saving at that income) in interest-investment space; their intersection defines the “investment rate” of interest at that income level, and tracing how this investment rate moves as income rises (depending on whether SS or CC shifts further) traces out the generalized IS curve (pp. 156-157). He also generalizes LL to reflect a monetary system in which authorities create money “up to a point” rather than holding M fixed, giving LL only a gradual upward slope (p. 157).

Generalized this way, Keynes’s theory “begins to look very like Wicksell’s.” There is one special case where it fits Wicksell’s construction exactly: if full employment means any rise in income immediately calls forth a rise in money wages, the CC and SS curves may shift right by exactly the same amount, making IS horizontal; in that case the investment rate becomes Wicksell’s “natural rate,” and whether the money rate is fixed at, below, or above it determines whether the economy has a stable price level, cumulative inflation, or cumulative deflation (p. 158). But Hicks stresses this is “only one special case.” With substantial unemployment (the “Slump Economics” Keynes is mainly concerned with), the sensitivity of the interest rate to income (∂C/∂I) is likely small and IS slopes downward; but when expectations are already inflationary and sensitive, ∂C/∂I can be large enough that rising income raises the investment rate, and it is then only an imperfectly elastic, rising LL curve that prevents the situation “getting out of hand altogether” (p. 158).

Hicks is explicit that his construction, though useful, is “a terribly rough and ready sort of affair”: the concept of aggregate “Income” is “worked monstrously hard,” since most of the curves in the apparatus are not really determinate unless something is also said about the distribution of income, not just its magnitude; what the curves actually express is better described as a relation between the price system and the system of interest rates than as a simple functional curve; and questions of depreciation and of the timing of the underlying processes have been neglected throughout (p. 158). The paper closes with Hicks’s own assessment of the book that prompted it: “The General Theory of Employment is a useful book; but it is neither the beginning nor the end of Dynamic Economics” (p. 159).