Notes
1 This is not to say that no Mainstream Keynesian economist has ever written a paper about unstable points of crisis in financial markets, and the few that did usually dealt with crises triggered by endogenous shocks.
However, before the outbreak of the recent global financial crisis in 2008, such papers were rare indeed, and the theory of efficient financial markets ruled the roost in mainstream financial economics. Moreover, much of the analysis of the recent financial crisis also relied heavily on exogenous shocks and not on endogenously generated instability. Hyman Minsky's interpretation of Keynes's theory of financial markets, which is adopted by many Post Keynesians, does deal with the theory of endogenous sources of financial instability.2 Hicks explained this property of Keynes's model via an "analogy" with the Walrasian static short-term multi-market equilibrium model. This is interesting because the Walrasian general equilibrium model exists in real time only when in a general equilibrium state. Trading outside general equilibrium is not permitted. When the economy does not have a marketclearing price vector, no real or calendar time passes as the "auctioneer" searches for the equilibrium price vector, so the system is always in general equilibrium in actual time. If trading at a non-equilibrium price vector were to take place, the distribution of wealth would change, causing the general equilibrium position to change. An analogous argument can be made about changes in expectations (and confidence and balance sheets) in out-ofequilibrium positions caused by the conflict between plans and outcomes. They will shift both curves.
3 Nothing in the model changes qualitatively if we assume that the money supply is moderately responsive to changes in the interest rate.
4 Hicks's critique of the LM curve does not apply to the IS curve.
If firms know the future states of the economy with certainty, they will be more - notless - likely to buy capital goods when the expected profit rate exceeds the long-term interest rate.
5 It does not take as long for a precipitous decline in security prices to destroy confidence in the expectation of continuation of a long-term market boom as it does to create optimistic expectations of a real-sector boom in the midst of a long depression.
6 This is an assumption in IS/LM models. In reality, it does not have to be valid in periods of either rapid expansion or rapid contraction where expectations and confidence are in a state of flux.
7 Keep in mind that YLREXt is a proxy for the mec or expected rate of profit on capital investment.
8 Note that a fully specified dynamic mathematical model could incorporate neoclassical risk if the expectation- and confidence-generating functions were assumed to be stochastic. But the model could not incorporate Keynesian uncertainty because the equations of the model could be used to calculate probability distributions of future outcomes that agents could use to make optimal choices.
9 The same logic can be applied to the impact of real-sector distress on the willingness of banks to make loans.
10 In a model with short-, medium-, and long-term interest rates, short-term rates would increase least because they are least affected by potential capital loss. They can be sold at par before much time passes.
11 For a defense of this assertion, see Colander and Landreth (1996).
20