One must acknowledge that estimating the long run is essentially a much more complicated procedure than OLS and yet because the results often end up with a single equation for the long run it looks much simpler. In fact, Engle and Granger 1987 article gave the impression that it was that simple and that OLS could be used. The additional idea that economic theory was essentially only about the long run added to the emphasis on the long run. The Johansen estimator tends to lead one to thinking that there is something almost magical about the long run that is worth studying on its own. Many papers that use the Johansen estimator have often left the problem after estimating the long run, sometimes blithely giving lip- service to how one shouldnow go on to making a full model. Peter Phillips emphasised the view of many that estimating the long run is only a stepping stone to a more sharply specified, economically interpretable, system of equations.
We are moving into a new phase in our understanding of these issues. Questions about the identification of the system are being raised but it seems unclear as to how specifying a separate system of long-run equations affects the identification of an economic system. For a while it seemed as if one could analyse the long run on its own and that economic policy should only aim at controlling the long run. However there is no causal direction in a cointegrating vector, only a co-relation over time. The question of control in nonstationary systems needs to be addressed carefully for the new econometric understanding to be of use to policy-makers.
ANDREWS, D. W. K. ( 1991), "'Heteroskedasticity and Autocorrelation Consistent Covariance Matrix Estimation'", Econometrica, 59, pp. 817-858.
ENGLE, R. F. and C. W. J. GRANGER ( 1987), "'Co-integration and error Correction: Representation, Estimation, and Testing'", Econometrica, Vol. 55, No. 2, pp. 251-76.
HARGREAVES, C. P. ( 1991), Australian Long-Run Money Demand, mimeo.
HARGREAVES, C. P. (ed.) ( 1992), Macroeconomic Modelling of the Long Run, Edward Elgar Publishing Ltd, Aldershot, UK.
JORESKOG, K. G. and D. SORBOM ( 1981), LISREL V: Analysis of linear structural relationships by maximum likelihood and least squares methods, Research Report 81-8, Department of Statistics, Uppsala, Sweden.
JUSELIUS, K. and C. P. HARGREAVES ( 1992), 'Long-Run Relations in Australian Monetary Data', Chapter 10, pp. 249-86, in Hargreaves ( 1992).
KWIATKOWSKI, D., P. C. B. PHILLIPS, P. SCHMIDT and Y. SHIN ( 1992), "'Testing the Null Hypothesis of Stationarity Against the Alternative of a Unit Root: How Sure are We that Economic Time Series Have a Unit Root?'", Journal of Econometrics, 54, pp. 159-78.
PHILLIPS, P. C. B. and B. E. HANSEN ( 1990), "'Statistical Inference in Instrumental Variables Regression with I(1) Processes'", Review of Economic Studies, Vol. 57, pp. 99-125.
PHILLIPS, P. C. B. and M. LORETAN ( 1991), "'Estimating Long-Run Eonomic Equilibria'", Review of Economic Studies, Vol. 59, pp. 407-36.