Exogenous shocks, stability, and comparative statics in

The General Theory and in IS/LM models

Let us first compare and contrast the effects of a substantial unexpected negative exogenous shock to investment spending on equilibrium income in the textbook IS/LM model and in The General Theory.

The direction of the initial impact on the equilibrium value of Y is con­sistent with the argument in The General Theory. However, Keynes's insist­ence in chapters 13, 15, and 22 of The General Theory that a sharp unexpected decline in income and employment is likely to be met by a sharp rise in r that exacerbates the downturn has been theorized out of existence in the IS/LM model because it assumes there is no change in expectations and confidence. In this very important sense, the IS/LM model is an anti­Keynes model. It eliminates the possibility of dysfunctional or "insane" gambling casino financial markets such as those of the late 1920s and 1930s whose importance was stressed by Keynes.

This decline in r will minimize the size of the fall in equilibrium income caused by the exogenous decline in investment in a manner similar to the reaction of r to a negative exogenous shock in the classical model. This is the result of a decline in the transactions demand for money, which increases the demand for bonds and thus lowers the current interest rate (while leaving the expected interest rate unchanged).

Of course, in the IS/LM model, the ultimate result of the shock is a lower level of equilibrium income, which clearly differentiates it from classical theory. It is worth repeating that the IS/LM analysis may not be unreasonable if we are evaluating the very-short-term effects of modest changes in the determinants of AD.

Now compare this IS analysis with the outcome of the effects of a nega­tive shock to investment spending within the framework of The General Theory. Given that expectations and confidence are endogenous, any sub­stantial shock to any section of the economy, if it is substantial, is likely to set off a dynamic and potentially unpredictable reaction that may or may not have a final resting place. The exact dynamics of the process cannot be determined unless the theorist is willing to fully specify the mathematical properties of the functions that generate expectations and confidence, but Keynes sketched out the logic of the process. Two aspects of Keynes's ana­lysis are especially relevant.

First, in Keynes's theory, the permanent fall in I and Y created by the negative investment shock should cause Y_LR^EX_t and the expected rate of profit or mec to decline. If the shock was unexpected, as is presumed, CF_t would decline as well. This would create an induced second leftward shift in the IS function as traditionally specified. The induced shift might itself cause further changes to expectations and confidence that would lead to yet a third leftward shift in the IS function - and so on.

Second, a significant and unexpected deterioration in conditions in the real sector is likely to cause financial investors to lower their expectations of future corporate profits and future bond and stock prices. They are also likely to become less confident in their ability to forecast financial asset prices accurately. Moreover, if the downturn is substantial and the financial system fragile, it could trigger debt defaults.

The increase in the interest rate caused by the induced LM shift will cause AD and Y to fall further, accelerating the downturn. This may cause expectations of future income and profits to decline yet again and/or confidence in the ability to forecast the future to fall, both of which will cause yet another leftward shift in the IS curve. In contrast to Mainstream Keynesian and classical theory, when serious trouble develops in the real sector, Keynes warned us that financial markets are more likely to worsen the effects than they are to help eliminate them. We know that Keynes had the catastrophic dynamic interaction between real and financial sectors in the USA in the late 1920s and early 1930s in mind when he was writing The General Theory. Yet this destructive dynamic is ruled out by assumption in the standard IS/LM model that, again, is supposedly the sole interesting model contained in The General Theory.

In the very unlikely and un-Keynesian event that both firms and finan­cial investors had foreseen the full effects of an unexpected initial "shock" to either sector and adjusted their expectations to it, it might seem that the ultimate impact of the shock will be new short-run equilibrium positions for Y and r as in the comparative-static exercises in mainstream macro textbooks. Yet even then, since the original shock itself was not anticipated in either market, there would be a decline in the confidence with which expectations are held in both markets, which itself would lead to an induced downward shift in both IS and LM curves.

A digression 283 cause another change in expectations and/ or confidence and thus a third shift in the curves.

The reader should understand that I am not conjuring up this inter­sectoral dynamic interaction based only on arcane hints offered in The General Theory. In chapter 22, Keynes clearly stated the problem, which can be explained in IS/LM terms. The sharp decline in the mpc will cause a fall in the IS curve, and this will trigger "a sharp rise in liquidity preference" that will cause an equally sharp upward shift in the LM curve.

The dismay and uncertainty as to the future which accompanies a collapse in the marginal efficiency of capital naturally precipitates a sharp rise in liquidity preference and hence a rise in the interest rate. Thus the fact that a collapse in the marginal efficiency of capital tends to be associated with a subsequent rise in the interest rate may ser­iously aggravate the decline in investment.

(CW 7, p. 316)

Keynes strongly emphasized the interconnectedness of disequilibrium processes in the real and financial sectors in his 1937 defense of The General Theory. He argued that the spread of pessimism or optimism from financial actors to business firms and from the latter back to the former was quite likely to occur. This makes sense: corporate leaders and the top executives of large financial institutions move in the same business and social circles, interact with and influence one another, have access to similar formal and informal sources of information about economic developments, and tend to hold similar views of the state of the economy. Thus, movement in either the IS or the LM curve, if it is significant and unexpected, will cause shifts in the other curve. Keynes said there is:

[no] reason to suppose that the fluctuations in one of these factors [investors' or lenders' expectations and confidence] will tend to offset the fluctuations in the other [entrepreneurs' expectations and confidence].

(Keynes 1937, p. 118, emphasis added)

He concluded:

It is not surprising that the volume of investment, thus determined, should fluctuate wildly from time to time. For it depends on two sets of judgments about the future, neither of which rests on an adequate

or secure foundation - on the propensity to hoard [or liquidity prefer­ence] and on opinions of the future yield of [real] capital assets.

(Keynes 1937, p. 118)

Neither the standard IS/LM model nor more sophisticated Mainstream Keynesian theory incorporates Keynes's vision of the potential for sub­stantial and even extreme financial and economic instability that was a major focus of The General Theory and of his 1937 defense of that book.