In The General Theory, Keynes presented his theory of financial markets in two parts.
He first analyzed the stock market in chapter 12, where he argued that because the future is unknowable in the present, expectations of future stock price movements can only be formed through the use of inherently unstable social, behavioral, and psychological conventions.
This was a key reason why the US stock market in the 1920s and early 1930s turned into an "insane gambling casino." He also argued in this chapter that capital investment spending is strongly affected by the movement of stock prices, making investment itself an inherently unstable variable. Keynes concluded that in the event of a significant and rapid fall in income, employment, and the rate of profit, stock prices were likely to fall rapidly, exacerbating rather than helping reverse the downturn. The stock market has potentially destructive disequilibrium dynamics.Having established this result, he moved on to analysis of the bond market and the behavior of the long-term interest rate in chapters 13-15. Since these chapters come after chapter 12, they assume the reader already understands Keynes's ideas about uncertainty and the properties of conventional expectations formation. He used these ideas to show that interest rates are likely to rise in the face of a sharp economic decline, as they did in the early 1930s - a result in direct conflict with classical theory and with IS/LM models as well.
I have reversed Keynes's order of presentation and turn first to his analysis of the bond market because he had spent the 1920s attacking the claim made by classical theorists that falling interest rates will always restore full-employment equilibrium in response to a substantial negative shock to demand. He did not spend substantial time trying to tie stock prices into his emerging macro theory until the early 1930s.
We begin with chapter 14, which presents and attacks the classical theory of the determination of the interest rate.
Classical theory has a flow equilibrium model in which the interest rate balances the flow of saving and investment at full-capacity income and output, a theory that assumes markets always clear in equilibrium. It also at least implicitly assumes that the stochastic future is knowable in the present.Classical theory:
has regarded the rate of interest as the factor which brings the demand for capital investment goods and the willingness to save into equilibrium with one another... [T]he rate of interest necessarily comes to rest under the play of market forces at the point where the amount of investment at that interest rate is equal to the amount of saving at that rate.
(CW 7, p. 175)
But, Keynes insisted, this could only be true "if the level of income is assumed to be given" or unchanged by shifts in the investment and savings functions (CW 7, p. 178). Full-employment equilibrium is the only possible equilibrium in classical theory. If, for example, there is a negative exogenous shock to the investment function at full employment, its effect on the interest rate as captured in classical theory is a reasonable approximation to reality if and only if the full-employment "level of income is assumed to be given" or constant throughout the disequilibrium process. In other word, this disequilibrium process proceeds almost timelessly while AS and income remain at their full-employment levels.
For the classical theory. assumes that it can proceed to consider the effect on the rate of interest of (e.g.) a shift in the demand curve for capital, without abating or modifying its assumption as to the amount of the given income out of which savings are to be made. The independent variables of the classical theory of the rate of interest are the demand curve for capital and the influence of the rate of interest on the amount saved out of a given income.
(CW 7, p. 179)
The key point here is that the classical conclusion that a negative shock to investment demand will cause interest to fall by the amount needed to restore the economy back to full employment turns out to be a tautology; full-employment production (AS) can be shown to be stable in the event of negative demand shock (AD) only by assuming its level never changes in response to that demand shock.
Let us examine the classical argument about the determination of the interest rate in some detail. We have just seen that in classical theory a negative shock to AD at full-employment equilibrium creates an excess AS of goods and services. This immediately leads to falling wages and prices and then to falling real wages. Falling real wages are assumed to keep AS perpetually at or near its full-employment level. However, if AD was to remain below AS for a long time when unemployment fell, unwanted inventories would pile up and workers would be laid off again. Thus, classical theory must provide an explanation of the forces that
Disequilibrium processes in bond markets 223 cause AD to quickly rebound back to its pre-shock level where it equaled full-capacity AS.
What are the out-of-equilibrium forces in the classical theory of financial markets that ensure that this AD rebound takes place? The answer is that an efficient bond market responds quickly to a negative AD shock by lowering the long-term rate of interest. This causes both investment spending and consumption spending to increase until their sum, which is AD in the simple model, is again equal to full-capacity supply.1 The classical theory of efficient bond markets thus plays a crucial role in the out- of-equilibrium dynamics that ensure a return to full employment after a negative shock to demand. Efficient bond markets are an essential foundation of Say's Law.
The classical theory of how changes in the interest rate keep AD approximately equal to full-employment output is, according to Keynes, roughly as follows. Income, Y, is identically equal to the value of output, AS. It is assumed that flexible wages perpetually hold AS near ASf (fullcapacity supply). But what causes AD to move back to equal ASF in the face of a negative shock to investment spending? The classical answer lies in the bond market.
The classical Quantity Theory of Money says that because bonds pay interest and money does not, people hold the minimum amount of money needed to finance transactions.
This means that all household saving (above the minimum needed to augment money holdings for transactions purposes) automatically flows into the bond market as an assured source of funds to finance business investment. Saving is therefore the demand for corporate bonds (BD). Classical theory assumed a time-preference theory of saving and consumption, with S = S(r, Y) and C = C(r, Y).2 The higher r, the more of our income we save and the less we consume. So BD = S(r, Y). If we assume with the classicists that all investment is funded by bond issues, the bond supply will be determined by the needs of business to finance investment: BS = I (r, Y).Since classical theory assumed that factor prices change quickly enough to hold AS approximately equal to YF (full-employment income) even in the face of negative shocks to demand, for purposes of analyzing financial market out-of-equilibrium processes, Y is treated in classical theory as an exogenous constant equal to full-employment Y or YF. Thus, even in the face of a negative shock to AD, we can write S = S(r: YF) and I = I(r: YF), where YF is an exogenous constant.3 The interest rate is thus the variable that regulates the balance between S and I or between BS and BD given that Y remains constant at Yf in the process.
Assume that the expected profit rate deteriorates. This is an exogenous negative shock to the investment demand function. Holding Y constant at YF, investment spending will decline at every level of the interest rate. At the original interest rate, S (the demand for bonds) will be greater than I (the supply of bonds). This will cause the price of bonds to rise and the
interest rate to fall. A declining interest rate will increase consumption spending and cause investment spending to rise relative to its value at the original interest rate. The interest rate must continue to fall until S declines and C and I increase by enough to restore equality between AD and YF again.
Since Y never changes by assumption (that is: ∆Y = ∆C + ∆S = ∆C + ∆I = 0), the interest rate will fall with almost infinite speed to cause ∆S to equal ∆I, and ∆C to equal -∆I. The rise in consumption spending equals the fall in investment spending so that AD again equals AS at Yf.4
It is thus the superefficiency of both labor markets and financial markets that gives classical theory its marvelous out-of-equilibrium stabilityrestoring processes. Full employment is not only stable, it is almost instantaneously stable. AS did not decline in response to the drop in AD (caused by the decline in investment) because wages and prices moved quickly to make the appropriate adjustments to AS. Then, given this rigidity in AS, the bond market moved at lightning speed to lower r, forcing I and C to rise and AD to move quickly back to equal YF. This comes very close to the auctioneered Walrasian general equilibrium model in which prices move the system back to general equilibrium in the aftermath of a shock before any actual or calendar time passes.