Tuesday, March 22, 2011

Catalan on an Internal Review of Economics

Jonathan writes:

"I think it’s time to ignore the role of the state [the role that politics has in the views economists hold], if it has any, and instead focus on introspection. It would be worthwhile to have some kind of organization activity, where the foundations of different theories are revisited, and where the science recovers some of its backbone (which, to me, is the coordination process). Without this type of internal review, it seems to me that economics will continue to grow in multiple directions, without any common basis, and the science will proceed down this road of disorganization and inaccuracy.

The differences in theory are real, not politically driven. Whatever arguments there are on the policy level, they stem from these differences in theory. It’s worthless to assess the political environment, and ignore the theory — if you’re looking to “fix” the profession, pointing fingers is not the way to do it
."

The review itself doesn't seem entirely necessary to me because I think I diagnose the problem differently. The problem with economics is not that we have multiple theories, it's that we've convinced ourselves that these theories are mutually exclusive. In the Austrian school, changes in the interest rate change the capital structure, which changes how resources are used. In Keynesianism, changes in the interest rate drive a wedge between the cost of capital and the returns on capital, which changes how resources are used. RBC theory posits real shocks that influence resource use. Minsky suggests that behavioral instability in financial markets introduces cyclical behavior. None of these ideas are mutually exclusive. What makes them mutually exclusive - what makes them appear "disorganized" to Jonathan - is that we've convinced ourselves that you can only agree with one of them. This is what bugs Russ Roberts so much too - he thinks disagreement is a sign that we are not scientific. Instead, I think it's a sign that we are just contentious people who disagree when we don't have to.

What economics does need to do is become more like biology. We are dealing with a very complex system where lots of forces are acting simultaneously. Our disagreements are not like those of Lamrack and Darwin, where we disagree on the fundamental processes involved. Austrians may talk about it slightly differently than the rest of us, but we all agree on subjectivism, marginalism, price coordination, and market efficiency. That's what makes economics what it is, just like evolution by natural selection makes evolutionary biology what it is. We are well past the fundamental disagreements. Our disagreements are more of the punctuated equilibrium vs. gradualism sort. We share a common foundation, but there are major disagreements in the way that that foundation is understood to unfold. We have a complex subject, just as evolutionary biologists do. It would be nice to resolve these sorts of disagreements conclusively. One day we presumably will and one position will be banished. But for the time being, the disagreement persists. Most of our disagreements are even more minor than that. For example, while our theories of interest rate determination represent a big disagreement (albeit with a common core of what we agree on), either a loanable funds, liquidity preference, or hybrid theory of the interest rate can accommodate Austrian theorizing about the capital structure. Austrians may find it hard to convince people that it is important, but that's OK. Nobody is obligated to consider something important just because another economist thinks it is. People have to be convinced!

7 comments:

  1. The problem with economics is not that we have multiple theories, it's that we've convinced ourselves that these theories are mutually exclusive.

    In a not entirely unrelated point, I've been tearing my hair out while having an email conversation with an anarcho-capitalist friend of mine this week. Among various other things related to the role of the state vs individual autonomy, we've been discussing natural monopolies.

    I complained to my friend that many treatments of this topic by (certain) Austrian writers completely misses the point, because they tend to focus on an industry as a whole rather than the particular natural monopoly component at hand. In doing so, they offer up a straw man caricature of what mainstream economists believe and completely misconstrue the argument to their followers. I offered electricity grids as an example...

    I've read criticisms by Austrian writers where they dismiss the notion of natural monopolies in electricity markets, by citing examples where decentralised/liberalised power solutions have existed in various places around the world. However, that completely misses the point as it conflates the actual grid (i.e. the natural monopoly component) with everything else, not least competition among suppliers.

    Virtually EVERYONE agrees about the benefits of liberalising electricity markets in terms of buyers and suppliers. (Finer points of regulation notwithstanding...) It is, however, the high-voltage national grid itself that is the issue of contention. Or, as I wrote:

    "If people want a competitive electricity market on any decent scale, the fact that the transmission grid is a natural monopoly will require for this key component to be provided by the state and/or heavily regulated (e.g. rate-of-return) to protect consumers from monopoly profits."

    My friend responded, using an amazingly flawed analogy, that:

    "To try and separate the two – the grid and the actual electricity - for the purpose of economic analysis isn’t valid. It would be like arguing that a coke and its can or the bottle that it is delivered in isn’t part of the final product, and that because the state can somehow manufacture coke cans most efficiently, no one else should produce them, enforcing a decree that itself or another producer of its choice has the ‘natural monopoly’ right to produce it[...] It is exactly your type of reasoning that leads to these gargantaun public utilities and 'natural monopolistic' industries that do not most efficiently satisfy consumers and stifles competition and hence does not provide cheaper and more efficient services for all."

    I don't even know where to begin...

    Apologies for ranting, but hopefully you can see the source of my frustration.

    ReplyDelete
  2. PS - I like Jonathan's proposal, although I sympathise with your comments that we often construct artificial disagreements where there are none.

    An additional problem, of course, lies with the method of verification. Problems with data aside... Certain epistemologies are fundamentally at odds with other (cf. strains of praxeology versus mainstream invocation of empirical testing). As long as these differences persist, you will cannot arrive at satisfactory conclusions because each group will hold the other's "evidence" as de facto invalid.

    ReplyDelete
  3. Eh? Daniel, is there a reason why my first comment has disappeared? Too long?

    ReplyDelete
  4. On the natural monopolies - it's been a while since I've thought about those issues, but even to the extent that your friend is right I agree he misses your point.

    So presumably some day what was a "natural monopoly" won't be a natural monopoly before. The division of labor is limited by the extent of the market - it's possible to conceive of natural monopolies turning more competitive. Anyone that advocates regulation of a natural monopoly (it seems to me) oughta concede that. That doesn't guarantee, though, that threat of entry and that sort of thing is going to guarantee competitive pricing.

    His example is obviously absurd. I'm not sure why he thinks you can't separate coke cans from the coke itself. Is he under the impression all of this is made in-house?

    ReplyDelete
  5. Ya, look... I'm certainly not saying that natural monopolies automatically persist over time; particularly in the context of changing technology, or recouped financing costs. (I'd be amazed if anyone made this claim.) I've written a number of comments on this blog to the effect of unbundling State assets to leverage competitive efficiencies... However, on this specific issue, I think the argument is perfectly clear. Moreover, it's important to separate financing costs from operating costs.

    Liberalisation of an electricity market typically means competition on all aspects, except operation of the grid. Why? Because the transmission system is characterised by increasing returns, etc, which means that it's inefficient for more than one firm to operate separate grids.[*] You would see a tendency towards market consolidation and ultimately arrive at a single provider[**], charging monopoly prices to competing suppliers/consumers who wish to use the transmission system.

    Regulation of the grid operator (via rate-of-return, price-caps, etc) is thus about preserving the competitive efficiencies created by market liberalization, not undoing them. Sorry to hijack your thread, but I honestly don't see what's so hard to understand about this.
    :'(

    [*] At the risk of being pedantic... There can be exceptions under peculiar circumstances (e.g. radial connections), but these remain very unusual.
    [**] This is exactly what was happened in places like Australia that experimented with (merchant) competition in the area of power transmission.

    ReplyDelete
  6. Ya - you're more talking about grids very specifically. Network effects in other words.

    I was opening up generally to the idea of natural monopolies as a question of optimal firm size.

    I think a lot of these people think of these two separately: http://en.wikipedia.org/wiki/Natural_monopoly#Network_effects

    So I was just throwing your stubborn friend a bone by noting that cost curves shift and what once was an industry characterized by such scale economies as to support a natural monopoly at one point may not be in the future.

    But network effects of the sort you note only get stronger with size.

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.