The Lucas critique comes back with a vengeance in DSGE models

from Lars Syll

Both approaches to DSGE macroeconometrics (VAR and Bayesian) have evident vulnerabilities, which substantially derive from how parameters are handled in the technique. In brief, parameters from formally elegant models are calibrated in order to obtain simulated values that reproduce some stylized fact and/or some empirical data distribution, thus relating the underlying theoretical model and the observational data. But there are at least three main respects in which this practice fails.

First of all, DSGE models have substantial difficulties in taking account of many important mechanisms that actually govern real economies, for example, institutional constraints like the tax system, thereby reducing DSGE power in policy analysis … In the attempt to deal with this serious problem, various parameter constraints on the model policy block are provided. They derive from institutional analysis and reflect policymakers’ operational procedures. However such model extensions, which are intended to reshape its predictions to reality and to deal with the underlying optimization problem, prove to be highly unflexible, turning DSGE into a “straitjacket tool” … In particular, the structure imposed on DSGE parameters entails various identification problems, such as observational equivalence, underidentification, and partial and weak identification. 

These problems affect both empirical DSGE approaches. Fundamentally, they are ascribable to the likelihoods to estimate. In fact, the range of structural parameters that generate impulse response functions and data distributions fitting very close to the true ones does include model specifications that show very different features and welfare properties. So which is the right model specification (i.e., parameter set) to choose? As a consequence, reasonable estimates do not derive from the informative contents of models and data, but rather from the ancillary restrictions that are necessary to make the likelihoods informative, which are often arbitrary. Thus, after the Lucas’s super-exogeneity critique has been thrown out the door, it comes back through the window.

Roberto Marchionatti & Lisa Sella

Our admiration for technical virtuosity should not blind us to the fact that we have to have a cautious attitude towards probabilistic inferences in economic contexts. We should look out for causal relations, but econometrics can never be more than a starting point in that endeavour since econometric (statistical) explanations are not explanations in terms of mechanisms, powers, capacities or causes. Firmly stuck in an empiricist tradition, econometrics is only concerned with the measurable aspects of reality, But there is always the possibility that there are other variables – of vital importance and although perhaps unobservable and non-additive not necessarily epistemologically inaccessible – that were not considered for the model. Those who were can hence never be guaranteed to be more than potential causes, and not real causes. A rigorous application of econometric methods in economics really presupposes that the phenomena of our real world economies are ruled by stable causal relations between variables. The endemic lack of predictive success of the econometric project indicates that this hope of finding fixed parameters is a hope for which there really is no other ground than hope itself.

This is a more fundamental and radical problem than the celebrated ‘Lucas critique’ have suggested. This is not the question if deep parameters, absent on the macro-level, exist in ‘tastes’ and ‘technology’ on the micro-level. It goes deeper. Real world social systems are not governed by stable causal mechanisms or capacities.

The kinds of laws and relations that econom(etr)ics has established, are laws and relations about entities in models that presuppose causal mechanisms being atomistic and additive. When causal mechanisms operate in real-world social systems they mostly do it in ever-changing and unstable combinations where the whole is more than a mechanical sum of parts. If economic regularities obtain they do it (as a rule) only because we engineered them for that purpose. Outside man-made ‘nomological machines’ they are rare, or even non-existant. Unfortunately that also makes most of the achievements of econometrics rather useless.

Both the ‘Lucas critique’ and the ‘Keynes’ critique’ of econometrics argued that it was inadmissible to project history on the future. Consequently an economic policy cannot presuppose that what has worked before, will continue to do so in the future. That macroeconomic models could get hold of correlations between different ‘variables’ was not enough. If they could not get at the causal structure that generated the data, they were not really ‘identified’. Lucas himself drew the conclusion that the problem with unstable relations was to construct models with clear microfoundations where forward-looking optimizing individuals and robust, deep, behavioural parameters are seen to be stable even to changes in economic policies. As yours truly has argued in a couple of posts — e. g. hereand here — this, however, is a dead end.