by Lucas critique and the modularity assumption Prima facie modular systems seem very special. But a number of authors suppose that modularity is the hallmark of causality … I have two objections to the usual claims about modularity. First, it is not a hallmark of causality. Recall the Phillips curve, a canonical example of a non-modular causal connection – one that, à la Robert Lucas, breaks down under attempts to manipulate the cause (inflation) to control the effect (unemployment). There are two standard responses to this failure. First: shaky equations just aren’t causal. To reply I turn to examples like the toaster and the carburettor where other conventional criteria – pushes, pulls, energy interchanges – argue that the connections are causal. The
Topics:
Lars Pålsson Syll considers the following as important: Economics
This could be interesting, too:
Lars Pålsson Syll writes DSGE models — a total waste of time
Lars Pålsson Syll writes Was MMT right about inflation? Yes!
Lars Pålsson Syll writes Brownian motion (student stuff)
Lars Pålsson Syll writes Keynes — en ständigt aktuell inspiration
Lucas critique and the modularity assumption
Prima facie modular systems seem very special. But a number of authors suppose that modularity is the hallmark of causality … I have two objections to the usual claims about modularity. First, it is not a hallmark of causality. Recall the Phillips curve, a canonical example of a non-modular causal connection – one that, à la Robert Lucas, breaks down under attempts to manipulate the cause (inflation) to control the effect (unemployment). There are two standard responses to this failure. First: shaky equations just aren’t causal. To reply I turn to examples like the toaster and the carburettor where other conventional criteria – pushes, pulls, energy interchanges – argue that the connections are causal. The lemonade-biscuit machine is an example where judgements about causality based on mechanical criteria go opposite to those from a manipulationist view. My verdict is pluralism: systems can be causal in different ways …
We want the causal processes in our day-to-day machines to be stable across reasonably hard use. The very shields that protect them make it hard to manipulate the internal causes separately. Also, typical machines, as well as many biological systems, are like the Lucas example: The causal connections under study depend on the stability of an underlying generating structure.
Modularity refers to the possibility of independent manipulability of causal relationships in a system. Trying to identify causal relations most economists today — especially when performing experiments — assume some kind of invariance or modularity, meaning basically that you can make an intervention on a part of a model without changing other dependencies in that model.
Modularity makes causal inferences made on the basis of ‘interventions’ stable. But although making causal inferences is not possible without making some kind of assumptions, you always have to argue why it is reasonable to make those assumptions. In the case of modularity that means you have to show that for the target system you are analyzing — the economy — it is possible to make ‘surgical interventions,’ ‘wiggle,’ or ‘manipulate’ parts of the system without changing other parts of the system. Since economies basically are interactionally complex open systems, it is de facto hard to find causes that are separately manipulable and show such invariance under intervention. Most social mechanisms and relations are not modular. Extraordinary claims require extraordinary evidence. So if economists want to continue to use models that presuppose modularity they have to start arguing for the reasonableness of it. As scientists, we should not merely accept what is standardly assumed. When is modularity a reasonable assumption and when is it not? That modularity makes it possible to identify causality in ‘epistemically convenient systems’ is no argument for assuming it to apply to real-world economies.
Coming up with models that show how things may possibly be explained is not what we are looking for. It is not enough.
We want to have models that build on assumptions that are not in conflict with known facts and that show how things actually are to be explained. Our aspirations have to be more far-reaching than just constructing coherent and ‘credible’ models about ‘possible worlds’. We want to understand and explain ‘difference-making’ in the real world. No matter how many mechanisms or coherent relations you represent in your model, you still have to show that these mechanisms and relations are at work and exist in society if we are to do real science. Science has to be something more than just more or less realistic storytelling or ‘explanatory fictionalism.’ You have to provide decisive empirical evidence that what you can infer in your model also helps us to uncover what actually goes on in the real world. It is not enough to present epistemically informative insights about logically possible models. You also, and more importantly, have to have a world-linking argumentation and show how those models explain or teach us something about real-world economies. If you fail to support your models in that way, why should we care about them? And if you do not inform us about what are the real-world intended target systems of your modelling, how are we going to be able to value or test them? Without giving that kind of information it is impossible for us to check if the ‘possible world’ models you come up with also hold for the one world we care about and in which we live — the real world.
The interventionist approach to causal inference has become more and more popular in social sciences during the last two decades. But most social systems are complex, evolving, contingent, dynamic, emergent, and genuinely uncertain. The theories and methods that build on the interventionist approach are not viable for those systems. As an unsubstantiated general assumption guiding causal analysis in social sciences, modularity should be abandoned. Other — more pluralist — methods and theories of causal inference and explanation are needed.