George Akerlof - Behavioral Macroeconomics and Macroeconomic Behavior (Part III : Effectiveness of Monetary Policy) lyrics

[III. Effectiveness of Monetary Policy] A central proposition of the New Cla**ical economics is that monetary policy, as long as it is fully perceived, can have no effect on output or employment. Perfectly foreseen changes in the money supply induce rational wage and price setters to raise or lower nominal wages and prices in the identical proportion leaving output and employment constant. This New Cla**ical hypothesis conflicts, however, with empirical evidence on the impact of monetary policy and the widespread popular belief in the power of central banks to affect economic performance. A major contribution of behavioral macroeconomics is to demonstrate that, under sensible behavioral a**umptions, monetary policy does affect real outcomes just as Keynesian economics long a**erted. Cognitive psychology pictures decision makers as “intuitive scientists” who summarize information and make choices based on simplified mental frames. Reliance on rules of thumb that omit factors whose consideration have only a small effect on profit or utility is an implication of such cognitive parsimony. In the wage-price context, simple rules cause inertia in the response of aggregate wages (and prices) to shocks—the exact “sticky wage/price” behavior that New Cla**ical economists had so scornfully derided. In the New Cla**ical critique, the inertial wage behavior hypothesized in the “neocla**ical synthesis” is irrational, costly for workers and firms, hence implausible. Behavioral economists have responded by demonstrating that rules of thumb involving “money illusion” are not only commonplace but also sensible—neither foolhardy nor implausible: the losses from reliance on such rules are extremely small. In joint work with Janet Yellen, I first demonstrated this result in the context of a model with efficiency wages and monopolistic competition. We a**umed that some price setters follow the rule of thumb of keeping prices constant following a shock to demand (caused by a change in the money supply). We showed that the losses to the “rule-of-thumb” firms from their failure to readjust prices following a change in the money supply are second-order (or small), whereas the impact on output of a monetary shock in this economy is first-order (or significant) relative to the size of the shock. We dubbed the rule-of-thumb strategies employed by firms with inertial price setting “near-rational” since the losses they suffer from their departure from complete optimization are second-order (or small). The logic of the key result—that near-rational price stickiness is sufficient to impart significant power to monetary policy—is simple. With monopolistic competition, each firm's profit function is second-differentiable in its own price so that the profit function is flat in the neighborhood of the optimum own-price. In consequence, any deviation from the profit-maximizing price causes a loss in profits that is small—second-order with respect to the size of those deviations. But if the deviations from the optimum of a large number of firms are similar—for example, if they are all slow to adjust their prices following a change in the money supply—then real balances (the money supply deflated by the price level) change by a first-order amount relative to a situation with fully optimizing price-setting behavior. This first-order change in real balances, in turn, causes first-order changes in aggregate demand, output, and employment. For example, suppose that the money supply increases by a fraction and a fraction of firms keep their prices unchanged. Each firm's losses, relative to fully optimizing behavior, are approximately proportional to the square of . If is 0.05, for example, its square is quite a small number, 0.0025, so the losses from price stickiness are apt to be small. However, a**uming money demand is proportional to income, the change in real output is first-order—proportional to . (With fully maximizing behavior by all firms, the change in the money supply leaves output unchanged.) Thus small deviations from complete rationality— indeed small and reasonable deviations from complete rationality—reverse the conclusion that expected changes in the money supply have no effect on real income and output. Rule-of-thumb pricing behavior takes many forms. For example, staggered price (wage) models, in which firms keep nominal prices (wages) fixed for a period of time, correspond closely to descriptions of price- (wage-) setting processes. In the Taylor staggered contract model, during each period, half of all firms set a nominal price which they maintain for the succeeding two period interval. A variant of the staggered contract model, due to Guillermo A. Calvo, a**umes instead that a fixed nominal price is reset at randomly varying intervals. New Cla**ical economists object to both renditions of the model, on the grounds that such price setting is not maximizing. Of course, they are right: instead of keeping nominal prices unchanged during a fixed interval, Taylor's and Calvo's firms would do better by establishing prices that vary within the interval in accordance with the firm's expectations of the money supply (aggregate demand). Such profitmaximizing behavior would again render money supply changes neutral. However, price-setting (wage-setting) strategies of the Taylor/Calvo type are near-rational: the small amount of nominal rigidity that characterizes these models is sufficient to allow monetary policy to be stabilizing, yet the losses relative to a strategy that varies prices within the pricing interval are second-order. There are many other forms of near-rational rule-of-thumb behavior that render monetary policy efficacious. Near-rational, rule-of-thumb models solve the great puzzle posed by Lucas regarding the effectiveness of monetary policy with rational expectations. New Cla**ical economics finds it difficult to explain more than a fleeting relation between money and output. The new behavioral economics, with a variety of plausible near-rational behaviors, yields a robust relation between changes in the money supply and changes in output.