Blog o’ Burns

Laidler, David. “The New-Classical Contribution to Macroeconomics.” Banca Nazionale Del Lavoro Quarterly Review, 1986.

More thoroughly than monetarism, new-classical economics has tried to define macroeconomic principles based on pre-Keynesian classical theory. Laidler argues that this key difference has yielded some overly stringent new-classical assertions, particularly its view that microeconomic behavior should be analyzed under the assumption of continuously clearing markets and strict rational expectations. But aside from these criticisms, he stresses the valuable contribution new-classical theory has made in terms of incorporating expectations into macroeconomic modeling.

Monetarism and New-Classical Macroeconomics

During the 1960s-70s, empirical evidence demonstrated that expansionary demand side policies yielded only temporary gains in output and employment, while inflation tended to rise continuously– a direct contradiction to the Keynesian model as it stood at the time. Monetarists responded by pointing out that: 1) Keynesians underestimated the impact of the quantity of money on aggregate demand and prices; and 2) the notion of a stable inflation-unemployment trade-off as shown by the Phillips curve was based on the assumption that the private sector was constantly fooled by changes in price level. Friedman and Phelps asserted that individuals act based on their expectations of the future, which can be derived from past experience. (Just as an example of this, I’m currently typing this blog on one of the library’s computers, which has a significant lag time between when I type and when the words appear on the screen. At first when I’d hold down the backspace key, I didn’t realize that I was pressing it for too long and would practically erase my whole post. I’m slowly adapting my typing so this doesn’t happen in the future!) They conceptualized the expectations-augmented Phillips curve and argued that any attempt to reduce the unemployment rate below its natural level would lead to higher inflation in the long run. Though these findings provided an alternative to Keynes, they didn’t necessarily challenge the theory and could easily be incorporated into it. Though the expectations-augmented Phillips curve could demonstrate an empirical generalization based on the assumption of expected inflation, monetarists could not explain how this outcome derived from the rational behavior of individuals at the microeconomic level.

Perhaps new-classical economics’ most enduring contribution to macro theory (according to Laidler) has been its version of the aggregate supply curve. Laidler demonstrates this by contrasting it to the Keynesian model. Keynesians assumed that sticky prices cause suppliers to reduce/increase since prices do not change fast enough for markets to clear. So an upward shift in aggregate demand would result in higher output rather than a sudden rise in the general price level. However, this belief in sticky prices could not be reconciled with the Phillips curve, which requires a rise in prices to correspond with an increase in employment. New-classicals provided an alternative explanation, under the assumption of flexible prices: An increase in aggregate demand will translate into higher prices; however, if the shift is unexpected, suppliers will assume that the higher price is unique to themselves, indicating an increase in relative prices. To take advantage of this relative increase, each supplier will produce more, causing aggregate supply to increase beyond its natural level. Thus a rise in general prices fools suppliers into lowering unemployment in order to increase production. However, if the rise in aggregate demand is expected (perhaps due to fiscal or monetary policy) along with a corresponding general rise in prices, suppliers will not hire more workers or increase output. The rise in aggregate demand will merely translate into higher inflation.

The Case for New-Classical Macroeconomics

The new-classical model claims to:

1) Explain past events and forecast future ones more accurately than previous models

Laidler refutes this claim, indicating that the Keynesian model (when modified by monetarism to incorporate expectations and the quantity of money) does a good job of explaining the events of the 1970s, in which prices were sticky and output changed in response to demand side shocks in the short run. This model explains the 1970s at least as well as any new-classical approach.

2) Yield insight into the nature of policy options and state the likely outcome of each choice

3) Have logical coherence and compatibility with other accepted doctrines

Laidler concedes that this third point in the strongest argument in favor of new-classical theory, which has tried to relate macroeconomics to microeconomics more than any prior model. To briefly go over this theory, if we assume that individuals operate in a world of perfect competition and imperfect information with respect to relative prices, supply and demand will depend on expected prices levels. Since people are rational, they will use all available information to form expectations that differ from the actual values only to the extent of a serially uncorrelated random error term. Thus, we can derive the aggregate behavior within an economy based on assumptions of individual behavior– or put the micro in the macro so to speak. New-classicals claim this approach is superior to Keynesian models which emphasize price-stickiness and thus require qualitative empirical laws, the parameters of which are defined by the data itself. Laidler counters this argument in his next section.

Empirical Evidence and ‘Free Parameters’

In the absence of free parameters, new-classical theory runs into trouble empirically when compared to what we know about microeconomics. For instance, the new-classical model gets rid of the free parameter linking money price changes to “excess demand” by theorizing that the Phillips trade-off reflects the elasticity of the supply of labor to changes in the real wage. However when this theory is tested, the results demonstrate that aggregate employment fluctuations seem systematically too large relative to inflation fluctuations to be treated as movements along a labor supply curve when the labor force confuses nominal wage changes for real changes.