Follow by Email

Monday, February 19, 2018

Rodrik: more on ideas vs interests

Dani Rodrik's weblog

Unconventional thoughts on economic development and globalization

My Photo


« Has Global Finance Reformed Itself More Than It Appears?Main

JANUARY 19, 2018

More on distinguishing ideas and interests -- an exchange with Peter Hall

My recent post on ideas versus interests elicited some comments from Peter Hall, my Harvard colleague who has done probably more thinking on this issue than any other scholar I know. With his permission, I am attaching these comments below, along with the brief subsequent exchange we have had.



A few quick thoughts inspired by your recent blog post on ideas and interests.  As you know, this is a topic that has long interested me.  These are rather cryptic thoughts (ars longa, vitae brevis) but I pass them along in case they are stimulating.

You are, of course, right to observe that interests are always interpreted (by ideas), i.e. they do not arise unambiguously from the material world.  And I think you are on the right track when you contrast the 'ex ante' account from interests with an ideas account of outcomes. 

But one might put even more emphasis (than I gather you do?) on the implication, which is that people always act based on both their ideas and their interests.  That is to say it is impossible to have perceptions of interest without ideas (and it is those perceptions of interest rather than the interests in themselves that motivate actors). 

Thus, claims that people are acting on their 'interest' are, in effect, claims that they are acting in line with some conventional set of ideas about what those interests are.  The classic example would be analyses that attribute to actors a set of interests that a conventional understanding of neoclassical economics would ascribe to anyone in their socioeconomic position.  The implicit claim of such analyses must be that those actors understand their position in that way, i.e. in line with these conventional ideas.

On my reading, this is what you mean when you associate interest-based arguments with a 'parsimonious' account of the attributes of the actors.  I do wonder whether parsimonious is the correct characteristic to highlight here (ie thinness vs thickness) since the ideas that underpin such action can be rather 'thick' (in the sense of depending on a relatively elaborate worldview).  The more important point, I think, is that these actors operate out of a worldview that can readily be seen as 'conventional' (in the sense of that term that it is widely-accepted as orthodox).

Now, there may be small sets of actors who in certain circumstances act against their 'interests' in the sense that they realize, by virtue of the ideas they hold, that they will lose something they value (material goods, power, etc.) by so acting.  And those actors might do that as a result of holding (other) sets of ideas, eg. of the sort associated with some sort of 'altruism'.  Soldiers who sacrifice themselves in battle might fall into that category.  However, I suspect that this is a very small category of people and, in many instances, as your argument intuits, such people could be said to have an unconventional view of their interest which dictates their action.

To continue then with the main account, if my view is correct, it becomes interesting to inquire about the role of ideas when there is (some kind of) contestation about precisely what is in the interest of the actors.  In your terms, these are cases in which ideas become salient to action And it turns out that is a relatively-common occurrence.  Thus, ideas often have influence over action, and the key problem is to explain (make claims about) why some ideas become influential in specific contexts while others do not.

With regard to that problem, it must surely be the case that a specific set of ideas (relative to other ideas) are more likely to become influential when they bear directly on the interests of the actors (understood as gains/losses of power or goods that the actor values).  Actors usually gravitate toward ideas that seem to them to serve their interests, understood in this stripped-down fashion. This is why interests and ideas typically bear together on action. 

What sorts of implications follow from this?

  • The Germans are probably not acting out of 'interest' independently of ideas. They are influenced by (Hayekian) ideas that tell them that their current posture is in the national interest as they construe it.  And, if we want to think of the latter as some sort of stripped-down 'material' interest, we have to bear in mind, first, that this conception of material interest is itself underpinned by an explicit set of ideas and, second, that there are other ways to interpret national interest and even material national interest.  For instance, policies that provoke a second Euro crisis might not ultimately be in that interest.  In other words, your initial point that all 'interests' must be interpreted by 'ideas' means that we cannot claim that interests trump ideas in instances such as this.
  • Although ideas are, on this view, important in all cases, we can detect, as you argue, instances that are distinctive by virtue of something about the prominence of the role that ideas play in them, such as the case of the Reagan tax cut. The issue is: what is distinctive about such instances?  I would argue it is not that ideas are somehow more important than interests in such instances.  After all, making the tax cuts was very much in the political interest of the Reagan administration.  What is distinctive is that there was contestation, with significant numbers of partisans on each side, about how to interpret (in this case) the economic interests of the populace.  It is the presence of contestation, rather than the importance of ideas, that distinguishes this case.
  • My bottom line is that the analytical way forward is not to ask: 'can we distinguish cases in which ideas were more important or influential from cases in which they were not?' but rather to ask: 'how might we best distinguish between situations in which ideas play a somewhat different role in the interaction between interests and ideas that underpins all action?'.

Don't hold me to this.  These are difficult issues and I find my views on them changing over time.  But I hope this is stimulating.



Dear Peter

Thank you very much for this. It is incredibly helpful, and I agree with much of it.

I am all for pursuing the agenda you set out at the very end – tracing out the different ways in which ideas affect interests and their interaction with prevailing ideational environment. But I would like to resist the formulation where interests are always subservient to ideas, which this approach presumes. There is a sense in which this is true, and you put it very well yourself in your note. A statement of the form, "the industry pursued its interest" must have some meaning, even though at the end of the day what we really mean to say is "the industry pursued its interest, as it perceived its interest to be."

We are often concerned with explaining changes. Why did an actor change its behavior? There is substantive difference, it seems to me, between two sort of explanations:

  • an explanation that relies on behavior in the context of an unchanged understanding of what the actor's interests are. The actor's utility function and its understanding of how the world works stay the same, but there are changes in the world it confronts. For example, the constraints it faces are altered.
  • An explanation that relies on a change in the actor's understanding of what its interests are. The actor's utility function changes or its understanding of how the world works is altered.

The distinction I would like to make is that case (a) is what many realists have in mind in IR or rational choice political economists tend to focus on. And I think it is OK to call these cases interest-driven cases.

Case (b) would fall in my ideational category.

Just like you, I am not sure about any of this. But we should try to figure out a way of thinking these issues more.





What you say makes excellent sense.  Focusing on cases of change is very promising.  And the distinction you draw between (a) and (b) a compelling one.  I agree that ideas do not trump interests.  The latter are just as important to be sure.

Of course, some of the most interesting cases are ones with features of both (a) and (b), i.e. constraints/opportunities in the world change and (partly by virtue of that) there is consideration, albeit not always adoption, of new ideas.  But the distinction strikes me as an excellent starting point.





Posted at 11:32 AM | Permalink



Aren't you focussing far too much on (explaining) behaviour of individual actors and failing to take into account that collective behaviour is complex (i.e. results from interactions between too many -agency imbued- actors to be analysable in that way) in which causality is not necessarily more important than intentionality, narrratives and competing discourses and in which path dependence makes analysis of behaviour being categorically caused by either x or y or both futile?
Agreeing on behaviour being driven by interests and/or ideas primarily implies agreement on a language and a perspective, that will subsequently performatively create social reality in its likeness.
If economists convince people that they act in accordance with ideas (and when they have a positive image of themselves: ideals) society will improve.
If economists convince people that they act in accordance with interests (and that serving one's own interest is only natural) society will deteriorate.
Let us please take our responsibility as economists to improve society!
Please see my "Economics as meant" presentation of a year ago in Leuven:

Posted by: WimNusselder | January 20, 2018 at 06:48 AM

Peter Dorman

I confess I'm at something of a loss in understanding the intensity of this exchange. It seems as though the two of you are both striving for an account framed in terms of "ideas" and "interests" as unitary terms, but surely there are very important distinctions we would want to make.

First, let's distinguish between ideas about what is in our interest from other ideas about what is right, advisable or entailed apart from our personal interest. If I believe it is in my personal interest to cheat you financially, we wouldn't want to conflate my idea about this interest with my idea about the rightness or advisability of carrying out the action, would we?

Second, the theory of ideology directs us to distinctions among ideas along a spectrum (or even multiple dimensions) of interest-impactedness. Even when people are acting, or claim to be acting, according to ideas other than those of their personal interests, they may be doing so indirectly to the extent they are attracted to these noble notions due to their interestedness. This is a complex, tangled topic, so I won't go further except to put in a reminder that the theory of ideology is about belief, not truth value. (Contemporary standpoint theory is utterly muddled about this.)

What about Germany? German political figures, pundits, the majority of the Council of Economic Experts, etc. generally justify their economic policies on the basis of ideas of the general good, not just what's good for Germany. In that sense, they pass the first filter: their ostensible motivating ideas are not narrowly ideas of their own national self-interest. The important question is whether they pass the second: to what extent is the current embrace of ordoliberalism reflective of the national interest of Germany given its reliance on current account surpluses and its position in the Eurozone? That's an important question, ultimately one that has to be explored empirically.

My own hypothesis, for what it's worth, is that structurally surplus countries generally display ideological predispositions for economic theories that stress saving and fiscal restraint. That's part of what it means to be a structurally surplus country.

Posted by: Peter Dorman | January 23, 2018 at 04:20 PM