Does macroeconomics have a right-wing bias?
…it’s really hard to make a DSGE model in which government policy plays a useful role in stabilizing the business cycle. By contrast, it’s pretty easy to make a DSGE model in which government plays no useful role, and can only mess things up. So what ends up happening? You guessed it: a macro literature where most papers have only a very limited role for government. …Thus, the conservative slant of modern macro comes not from the weight of evidence, but from the combination of publication bias and the inherent unwieldiness of the DSGE framework.
Update: The comments below make it clear that I should have explained what I think is interesting about this quote, and it’s got nothing to do with DSGE in particular. It is the general point that assumptions are often made for tractability, rather than realism, yet still influence our conclusions. It isn’t possible to control for the unrealistic assumptions; if it were we wouldn’t have made them. That means our conclusions will be biased by assumptions we’ve made only for convenience and we need to bear that in mind when considering the policy implications of our models. For example, if our model assumes perfect competition and our conclusions rely on prices adjusting then we might need to be a little sceptical.
Of course, DSGE’s are only designed to analyse monetary policy and given an idea of what impact monetary policy changes have in terms of stabilising the business cycle – structural government policy isn’t explicitly modeled and requires a different framework … this is common knowledge in the discipline
I think that’s exactly his point: government policy isn’t explicitly modelled so there is an absence of recommendations for government action in the macro discourse.
DSGE modeling isn’t “macroeconomics” – it is a modeling technique used to understand monetary policy given a certain form of preferences and structural government policy.
People who look at structural policy use different models, or there is also a growing literature with people trying to generalise the DSGE framework to allow for specific types of government action.
Acting like DSGE’s are akin to all macro models (which the aforementioned quote appears to be) is like saying that perfect competition is akin to all micro models – it is worse than making a poor strawman, it is just a completely false statement 😉
I don’t think anyone’s acting like that but when a particular class of model is considered canonical within the discipline then its application influences the discourse. Its weaknesses affect the type of conversations that people have. That happens whether the weaknesses are recognised or not.
Within macroeconomics, DSGE models are generally only used to analyse monetary policy – unless they have been specifically “fit for purpose” to analyse something else. In the most part, when people want to analyse the role of government they use different models – and anyone trying to use the government spending term in a DSGE model to do that would be looked at like they didn’t know what they were doing.
Macro literature that involves DSGE’s has nothing to do with government policy, making this critique entirely aimless.
There is another whole set of macro literature discussing government policy, if we are going to say that macroeconomists looking at fiscal policy have a “right wing bent” we should be looking at that literature – rather than aimlessly criticising an unrelated set of models.
As you well know, there are plenty of papers using a DSGE framework to examine both fiscal policy and the government’s potential role in monetary policy.
Nonetheless, you are avoiding the question raised by the quote, which is whether the assumptions we make for convenience have a significant influence on our eventual conclusions.
Ok, answer me this. How is saying that stabilisation policy is the role of monetary authorities conservative/right wing?
Does this literature suggest that there is a limited role of government for stabilisation policy – yes, yes it does.
But this quote:
“You guessed it: a macro literature where most papers have only a very limited role for government.”
Is completely false, because there is a significant amount of macro literature that discusses the role of government in a structural sense. The DSGE framework is, in of itself, of very little use here – any DSGE models used to discuss the role of fiscal policy tend to be augmented with something else, in order to discuss a specific “tendency” – and often in a way that is noted as explicitly inferior to other forms of analysis.
DSGE modeling, even the New Keynesian “synthesis”, is actually only a minority part of the macro literature (we are just hearing the most about it, because we are in the midst of a large recession) – I find the idea that because this literature focuses on the business cycle (which is its entire point) sensible, not some random conservative bias.
I have noticed that industrial economics says very little about stabilisation policy – does this imply that there is an “anti-inflation targeting bias” in industrial economics? You may think I’m joking, but this is truly an equivalent statement.
I can no doubt word this more concisely.
In the DSGE literature, which is a subset of macro literature, we are modeling the impact of monetary policy – given that government policy is exogenous. This is to help inform monetary policy.
If we want to discuss issues of government policy, then we use different models – as we often do when looking at tax and other structural issues. We can even model fiscal policy as stabilisation policy, given the central bank reaction functions we are able to model, with help from the DSGE literature – and sure enough there has been a lot of this as well.
However, just because the subset of a discipline focuses on answering a specific question doesn’t mean that the entire discipline has a “conservative bias”. To say this we instead need to say that there is “too much” focus on DSGE models, and “too little” focus on issues of fiscal policy – that is certainly not what the quote is suggesting, and is in itself a relatively big claim (although one I think would be interesting to look into).
“That’s all a sideshow to the central question of how assumptions made for tractability then influence our conclusions and policy recommendations. For example, you point to perfect competition as an analogy and I would argue that the assumption generates a significant bias in applied economists’ work that is often overlooked.”
The general point you raise here is a very important one – although I would say that this comes more from your thoughtfulness then from the quote mentioned above.
This brings us back to the very issue of identifying core and peripheral assumptions when looking at an issue of interest – you are stating that, sometimes when we look at an issue, we make implicit claims about core assumptions which are false … and we hide these by saying that they are “exogenous. This is dangerous, and wrong, and avoiding doing this is a central lesson that we gain from the study of economic methodology.
In the case of DSGE modeling, and its central goal to model the impact of monetary shocks, the treatment of government policy as exogenous appears to me to be a fair one.
Beyond this, the statement that using a DSGE model in this way has a “right wing” or “conservative” bias appears very odd – as DSGE models don’t give us policy relevant information about fiscal policy.
Now don’t get me wrong – the issue you raise in this last paragraph is very important. Often economists will claim a result, when it is really due to some CP assumption being inappropriate. And these sorts of issues should be tackled anywhere they are! In this specific case though, I think the criticism is unjustified – and has involved the role of DSGE modeling in economics being misattributed.
“On the substance of what you’re saying, you’re referring to a subset of the DSGE modelling done. Your critique is very fair for the problem you’re talking about but the DSGE framework has since been exported to other areas of macro, as I understand it.”
DSGE modeling only took off post-2005, and almost every paper I’ve read that uses DSGE models to discuss a non-monetary related policy issue does so by augmenting the model with something else – given that any discussion of fiscal policy within a typical DSGE model is empirically empty.
I’ve also noticed that these papers tend to come up with much more interventionist policy prescriptions than the standard literature in their fields – so if anything, I’d be more convinced that there is a left leaning bias in their implementation rather than a right leaning bias …
I accept your point about making sure we include all “core” assumptions in any model when answering a question – or else we could be implicitly assuming something inappropriate. This is stock standard good economic methodology – and the answer and the elements we assume (do we rely on one model that includes all core assumptions, or a set of models that test different tendencies given subsets of core assumptions?) should be something that is closely kept in mind. The realism of core assumptions is straight out of Maki.
However, I disagree completely that the make up of a standard DSGE model is proof of a right-wing/conservative bias – which is the entire thrust of what comes from the title of the post and the quote that has been pulled out.
I’m not an economist, but for me only the micro matters. The entire market need only be considered from the point of view of every single transaction that occurs between individuals, which renders the macro, and aggregates, and thinking a central planner can pull this lever to change any variable in an assumedly known direction, etc, absurd.
I always like micro when it ends with brewery, also.
The distinction between macro and micro is meant to be more akin to the type of question asked, rather than the general tools used – macro questions deal with a broader issues.