Critically discuss the following statement: “The Sargent and Wallace (1976) model of policy ineffectiveness has no basis in reality. It is of no practical or theoretical value to policymakers and economists alike.”
The Sargent & Wallace model (1976) produced the ‘Policy Ineffectiveness Proposition” which is viewed as a radical turning point for monetary theory and part of the ‘New Classical’ revolution that dominated policy during the 1970’s and 1980’s. Despite criticisms, it holds great significance as a benchmark model.
The model is built upon the Lucas supply function:
(1) yts = yn + (pt – t-1 pte) + u t
This stipulates the natural rate hypothesis that output can only deviate from its natural level by price forecasting errors or a random supply shock.
The money supply rule is given by:
mt = α + β (y* – yt-1) + εt,
Where α is a constant term, β is a parameter and y* is a target level of output.
And ultimately, output in the model is given by:
(3) y t S = y n + εt + ut
It can be seen from (3) that the parameter set by anticipated monetary policy has no effect on the behaviour of output. Only the unanticipated money shock, εt , will have effect.
The model is structured upon New Classical assumptions of rational expectations (RE), a Lucas supply curve and that only real variables matter. By substituting for more realistic assumptions, the policy ineffectiveness proposition would not hold.
RE is defined below:
t-1 Pt e = E (Pt / t-1)
RE contrasted with the backward-looking expectations assumption of the adaptive expectations model that dominated previous theory. With RE, an activist policy would be predicted by agents who would then revise wage and price expectations upwards, resulting in unchanged real variables. There is no money illusion and agents do not make systematic mistakes.
However empirical evidence suggests persistent expectational errors, seen by constant underestimation by agents of UK inflation (Carlson & Parkin, 1975). Friedman used expectational errors to argue against the short-run neutrality of monetary policy. However RE is widely accepted, shown in the impact of inflation forecasting by the Bank of England has upon expectations and its use within the Efficient Market hypothesis. However the acceptance is seen as, “necessary but not sufficient” (Spencer, 2009) for the validation of the PIP, as models that are fully consistent with the rational expectations hypothesis with more realistic assumptions, have taken precedence.
One such model, and a critic of the Friedman style ‘market-clearing’ assumption was Fischer (1977). He introduced short run wage rigidity, with agents making nominal contracts that lasted longer than one period. Monetary policy could change at higher frequencies than prices and wages, implying non-neutrality in the short run, Taylor proposed nominal rigidities in his model, with the inclusion of staggered wage contracts with similar results. The market clearing model seems distinct from reality, with real world lags. This assumption is credited by the Bank of England, who set a horizon for up to two years for achieving their inflation target and suggested adherence to a Taylor style rule. The Keynesian assumption is that the large unemployment seen throughout the world today is evidence that labour markets do not clear. The assumption of fully flexible prices is discredited by the Calvo model. Its inclusion of menu costs supports the fact that numerous imperfections within today’s economy stop people reacting to news immediately.
Hoover states if the symmetric information structure is removed, monetary policy does affect real variables. Grossman & Stiglitz (1980) state that agents would not pay the cost to become informed as under rational expectations no profit could be made. This leaves policy-makers with an informational advantage and the ability to affect real variables. Support for symmetric information structures is seen via the UK, where transparency is vital, thus information differentials not persisting for long.
In addition to invalid assumptions included within the model, it has been criticised for its exclusions. Econometric evidence suggests when assessing factors affecting output, exclusion of “monetary..policy would…create the greatest potential shortcoming” (Hutchinson & Glick). Shammout argued the impact of monetary policy upon interest rates, exchange rates & stock prices, instead of just prices, that can affect output. Money is seen as the only financial asset, excluding even government bonds. There is little evidence supporting its practical application, with early evidence by Barro (1977) deemed a “research failure”. Blanchard (2003) postulated the Mundell-Tobin effect of the ability of monetary policy to alter the natural rate of unemployment, with evidence in the evolution of European unemployment. Mishkin (1982) found both anticipated and unanticipated monetary policy has effect on real variables in the short run. The Quantitative Easing programme in the UK, seen to have helped unemployment, would be ineffective if the PIP held.
Despite criticisms, its importance within monetary policy cannot be underestimated. The influential Barro-Gordon model (1977) supported the model with the assumption that whilst output and employment were affected by unanticipated monetary policy, anticipated policy would have no effect on real variables. The ‘Real Business Cycle’ model confirmed policy ineffectiveness in a world without the market-clearing assumption. It has promoted widespread use of the RE hypothesis, equilibrium modelling and cemented the need for firm microeconomic foundations in macroeconomic policies (Snowdon & Vane). The New-Keynesian models are seen as emanating from the new classical challenge, in which Sargent & Wallace played a key role.
The Sargent & Wallace model significantly impacted upon monetary policy, although not as its creators anticipated. Modern economists generally accepted the New Keynesian approach of the long run neutrality of monetary policy, and its short run potency due to real and nominal rigidities. Critics argue that the model presents a simplified static world, of complete certainty with no relevance in the real world. However in the light of theoretical application, “unrealistic assumptions are in fact necessary in the formation of a good theory” (Gilbert & Miche) Thus although its modern practical use is negligible, its application within theoretical developments are vast.