Arnold Kling  

Regional Disasters: AS/AD vs. PSST

PRINT
Effect of Monopoly Unions on I... Your Taxes at Work...

Suppose that a natural disaster occurs in one region. How does this affect the rest of the world? In the aggregate supply and demand paradigm, it would seem that the rest of the world benefits from stronger net exports. In the paradigm of Patterns of Sustainable Specialization and Trade, it would seem that the rest of the world is hurt by the destruction of normal trade patterns.

Thus, PSST offers a different "prediction" than the AS/AD paradigm. However, the only way to test this prediction would be to run a controlled experiment, with the same world economy first given a "control" path and then re-started with the same initial conditions and given a path that includes a regional disaster. So, I offer this analysis not so much for empirical testing but as a way of highlighting the differences between the paradigms.

A regional natural disaster is presumed here to destroy part of the capital stock, including manufacturing plants, office buildings, and residential structures. We assume that the region soon begins to rebuild, drawing down saving and using aid from outside the region to pay for imported goods from outside the region.

Within the affected region, from the AS-AD perspective, the destruction caused by the disaster is an adverse supply shock. On the demand side, reduced saving is expansionary. The overall effect on output might technically be ambiguous, although it would seem reasonable to suppose that the supply shock is more important, so that output declines.

Outside the affected region, the rest of the world experiences no supply shock. Assuming that the affected region exports less and imports more, this raises aggregate demand in the rest of the world. Output goes up in the rest of the world.

The PSST paradigm would look at the situation differently. The disaster would disrupt patterns of trade, both inside the region and outside. The disruption would be larger inside the region. However, the effects would be adverse both inside the affected region and outside of it. In the rest of the world, relative to a no-disaster scenario, output would be lower rather than higher.

From the PSST perspective, the productive capacity of one region is important to the rest of the world as well. The rest of the world has established patterns of trade that rely on the specialized output and consumption patterns of the affected region. Losing those trade patterns will require an adjustment. The process of adjustment is costly and requires a significant period of time.


Comments and Sharing


CATEGORIES: Macroeconomics



COMMENTS (18 to date)
Daniel Kuehn writes:

I'm regularly confused by the inclusion of "vs." in these sorts of posts - "AS/AD vs. PSST", as if we must choose.

PSST is essentially a point about specialization and comparative advantage - about the way that goods and services are efficiently supplied. These are not concepts that are rejected by AS/AD approaches. PSST is the bedrock of the discipline. It seems to me all PSST implies is that disasters have AS impacts outside the region just as they have AD impacts that are worth talking about. It doesn't seem to be an approach that is opposed to AS/AD. It's a micro/macro distinction. Juxtaposing these two makes about as much sense to me as juxtaposing micro and macro.

A dude writes:

PSST includes everyone searching for useful things to do. A large scale distruction of infrastructure provides an obvious and proven "pattern", so less "searching" is needed, just repeat what was done in the past to built those houses, bridges, etc

Same effect as AS/AD

Gregor Bush writes:

"Outside the affected region, the rest of the world experiences no supply shock."

Suppose the rest of the world is completely obliterated. Is this a negative supply shock for the US? I would say that it clearly is. It would lead to much lower levels of productivity because the US is no longer able to benefit from the gains from greater specialization and division of labor that international trade provides. The same would be true, although to a lesser extent, if only Japan, the world’s third largest economy, were wiped out.

So I think in the AD/AS framework there is a negative supply shock for the rest of the world and the channel to the AS curve is through productivity.

Silas Barta writes:

@Daniel_Kuehn: Then why do Keynesian economists always (as soon as it's socially acceptable to do so) remark that a particular natural disaster will "help the economy" because of spurring job growth and aggregate demand? I see this all the time from Krugman et al the wake of e.g. hurricanes.

Why don't they see that such "GDP growth" in these situations no longer indicates improved economic health, but restoring the previous level of economic health due to rebuilding? When economists can show understanding of the limits of their own metrics, people like Arnold Kling can stop thinking they need to hear remarks like those in his post.

Daniel Kuehn writes:

Silas -

Often economists are asked "what is the impact on the economy going to be?". It's reasonable to ask economists that, just like it's reasonable to ask environmental scientists what the impact on the environment will be. The potential positive shock to production is a counter-intuitive result a lot of people don't think about, so naturally economists are going to highlight that. But you act like they're chomping at the bit to point this out. I'm not sure that's the case. "What will the impact be on the economy?" is always a question that gets asked by reporters. Economists naturally like counter-intuitive results, of course. That's a very common disposition among them. But usually they say these things because they get asked these things.

Larry Summers dedicated two sentences in an interview to the possibility that this will bolster GDP. Russ Roberts and Don Boudreaux have, by my count, already dedicated four blog posts totalling over 1600 words to what Summers said. I think the extent to which economists focus on this GDP result is wildly distorted by people who want to slander them over it.

Take a look at Krugman's "meltdown macro" post. He highlights two things - (1.) the international supply chains that this will disrupt, and (2.) the point that because of the liquidity trap one thing he's not worried about is the bond market.

That doesn't sound like the sort of response you're imputing to him. None of these guys promote GDP as a metric of total economic health. It's a production measure, period. If it is to be extended beyond that, perhaps we can remark on its relationship with employment levels (although that tie is getting more dubious lately). But nobody promotes it as a measure of general economic health. For decades economists have gone to great lengths to highlight the fact that GDP is not a measure of wellbeing.

Daniel Kuehn writes:

Silas -

Often economists are asked "what is the impact on the economy going to be?". It's reasonable to ask economists that, just like it's reasonable to ask environmental scientists what the impact on the environment will be. The potential positive shock to production is a counter-intuitive result a lot of people don't think about, so naturally economists are going to highlight that. But you act like they're chomping at the bit to point this out. I'm not sure that's the case. "What will the impact be on the economy?" is always a question that gets asked by reporters. Economists naturally like counter-intuitive results, of course. That's a very common disposition among them. But usually they say these things because they get asked these things.

Larry Summers dedicated two sentences in an interview to the possibility that this will bolster GDP. Russ Roberts and Don Boudreaux have, by my count, already dedicated four blog posts totalling over 1600 words to what Summers said. I think the extent to which economists focus on this GDP result is wildly distorted by people who want to slander them over it.

Take a look at Krugman's "meltdown macro" post. He highlights two things - (1.) the international supply chains that this will disrupt, and (2.) the point that because of the liquidity trap one thing he's not worried about is the bond market.

That doesn't sound like the sort of response you're imputing to him. None of these guys promote GDP as a metric of total economic health. It's a production measure, period. If it is to be extended beyond that, perhaps we can remark on its relationship with employment levels (although that tie is getting more dubious lately). But nobody promotes it as a measure of general economic health. For decades economists have gone to great lengths to highlight the fact that GDP is not a measure of wellbeing.

Philo writes:

"In the aggregate supply and demand paradigm, it would seem that the rest of the world benefits from stronger net exports." But also from weaker net imports; so why call this a net *benefit*? Does "the aggregate supply and demand paradigm" consider exports good, imports bad? Is it mercantilist?

Silas Barta writes:

@Daniel_Kuehn:

The potential positive shock to production is a counter-intuitive result a lot of people don't think about, so naturally economists are going to highlight that

But it's not simply counter-intuitive; it's _wrong_, and to the extent that economists fail to incorporate the economic damage Arnold Kling is referring to in their definition of a "good economy", they are truly lost.

Natural disasters are not a "positive shock" to production. To the extent that any metric is "improved" because of a disaster, it means that the metric diverges from what we are trying to measure in this scenario, not that the natural disaster was "actually good", or some blessing in disguise.

The economy is good if and to the extent that people are generally moving up in their preference rankings. If that's not happening, it shouldn't be reported as a "good economy", no matter what the metrics show. Obviously, you can't always know this, but when you can, you have to recognize the limits of the measure.

Now, Krugman may be acting better than usual in the light of this latest disaster, but its very routine for Keynesian-oriented economists to claim that a given disaster will help the economy, specifically on the basis of GDP or aggregate-demand reasoning. If you really doubt this, I would be glad roll out with a search showing how his happens over and over.

Yes, economists point out the shortcomings of GDP, but not in any way that allows them to think correctly about natural disasters.

Jacob AG writes:

@Philo

Krugman's point is not mercantilist per se. He argues that all that borrowing Japan will have to do to rebuild *could* pick up the slack in labor and capital markets: it could put excess savings to work, close the output gap, lower unemployment outside Japan, and thereby make productive use of resources which would otherwise be sitting idle. He is "terrified about the possible loss of life [and] nervous about the disruption of world production," but he sees a Keynesian multiplier in Japan's increased borrowing.

Kling is very skeptical of all this Keynesian stuff, and prefers to focus on the disruption of world trade. I worry that some of his readers will think that the AD/SD model and its proponents imply that *all* disasters are expansionary. Krugman, for what it's worth, hasn't suggested that; he's only said there is a silver lining.

Pierre writes:

As a practical example several Japanese camera factories have had to close, including Nikon and Cannon, which will lead to shortages and price increases on their products, which will, I predict, also raise the prices on some of their competitors products. Customers of Japanese photographic products, therefore, no matter where they are in the world, will presumably pay these higher prices, leading to reduced consumption elsewhere.

The counter argument is that the glaziers will be doing a gangbusters business.

Pierre

Pierre writes:

I also remember circa 1993 or 1992 when there was a large fire that destroyed a significant portion of the world's component supply for computer memory. Prices doubled per megabyte, from $50/Mb to close to $100/Mb, and stayed there for a significant period of time - a year or two IIRC. That definitely caused me to buy less computer than I would have liked to buy, in my case, by not buying a computer and holding on to my old one for a few years more.

Pierre

N. writes:

Surely, though, no one could argue that a natural disaster can't be a boon to certain market sectors? Or that reconstruction can't have hidden productive value?

If, for instance, you have a bunch of construction workers sitting on their duffs, I can see where they would stand to benefit rather immediately from a natural disaster... and when stuff is rebuilt it tends to be using the latest & greatest as opposed to the old 'legacy' material.

Furthermore, yes, as Bastiat would have it, the shoemaker may go broke elsewhere... but couldn't he also be forced into discovering new PSST that actually improves his bottom line?

I'm not advocating going around breaking windows, but it's called "creative destruction" for a reason, no?

Daniel Kuehn writes:

Silas -
re: "To the extent that economists fail to incorporate the economic damage Arnold Kling is referring to in their definition of a "good economy", they are truly lost."

Well yes... if economists do bad analysis the analysis will be bad. You think I'm going to disagree with you on this? The fact is we don't know what the impact will be on GDP. It's ambiguous. I would have thought the correct response in that case would be "the impact on GDP could potentially be positive because of rebuilding efforts, but it's unclear because this disrupts our productive capacity". Krugman seemed to be saying that. Am I missing something?

re: "it means that the metric diverges from what we are trying to measure in this scenario, not that the natural disaster was "actually good", or some blessing in disguise"

Well this is a different point entirely. I don't know of anyone who has argued that the disaster is good or a blessing in disguise. Do you?

Silas Barta writes:

@Daniel_Kuehn:

The fact is we don't know what the impact will be on GDP. It's ambiguous. I would have thought the correct response in that case would be "the impact on GDP could potentially be positive because of rebuilding efforts, but it's unclear because this disrupts our productive capacity". Krugman seemed to be saying that.

You're confusing separate issues here. Yes, the impact on GDP will probably be positive, but no, a natural disaster is not good for the economy (in the sense of "the economy" that people actually care about enough to regard it as a worthwhile discipline).

Do you understand the difference between the two? If not, you're in the category of lost economists.

I don't know of anyone who has argued that the disaster is good or a blessing in disguise. Do you?

I'm saying that economists have said that disasters have been good "for the economy". The problem is that they don't see the contradiction between "having that disaster really sucked" and "that disaster was good for the economy".

And yes, they say the latter routinely.

Silas Barta writes:

@N.:

Yes, certain sectors will benefit, but the economy as a whole will not. If workers were sitting on their duffs, it still would have been better for them to retool to some other production mode than for the disaster to happen just to keep them employed.

Yes, destroying old capital can be good on net, if people wanted to replace it anyway but couldn't afford demolition costs and the disaster happened in a way that focused the damage precisely on these structures, but that happens with vanishingly small probability. (It's like arguing that farm subsidies are good because the aliens that are secretly monitoring us will show up and give us free chocolate if we have *just* the right farm policy.)

If that's not the case, there is still a net economic loss once you factor in the replacement costs.

The same reasoning applies to the case of new PSST that someone discovers as a result of a disaster: yes, they could happen, but people have such serendipitous discoveries with or without disasters, and you can't predictably increase the net gain from such discoveries by deliberately crippling yourself.

Does that answer your questions?

Daniel Kuehn writes:

re: "You're confusing separate issues here. Yes, the impact on GDP will probably be positive, but no, a natural disaster is not good for the economy"

Well then what was your problem in the first place? This is all anyone has said!

I agree, and would have been able to say I agree sooner if I had realized that was all you were saying.

re: "And yes, they say the latter routinely."

OK... we're back to disagreement.

If they say it routinely, could you provide an instance where they don't restrict their attention to GDP/production/investment?

Silas Barta writes:

@Daniel_Kuehn: An economist is making the error I describe any time they claim (as Keynes often did, e.g. his lauding of the practice of building pyramids or burying money) that employing people for a useless or make-work task can help the economy.

Even if they were to carefully restrict their remarks to GDP or aggregate demand or whatnot, do you not see how bizarre that is? What people want to know about is general economic health, and and indicators that actually indicate this. Why would you report on a metric in precisely those times when it fails to be a useful indicator of what people want to know about, and the difference matters? That would be like a weatherman pouring water into his rain gauge and then reporting,

"Today our rain gage shows 5 inches."

Yes, everything he's said is true, and maybe he understands the limits of that measure, but he's fundamentally failing to address what people *think* he's talking about.

(But they probably *don't* understand the corresponding distinction, as you always find Keynesians wishing people would "do their part" to bring us out of the recession by going deeper into debt.)

Joey Donuts writes:

@Daniel Kuhn

Often economists are asked "what is the impact on the economy going to be?". It's reasonable to ask economists that, just like it's reasonable to ask environmental scientists what the impact on the environment will be. The potential positive shock to production is a counter-intuitive result a lot of people don't think about, so naturally economists are going to highlight that.

The idea of a positive shock to production isn't counter intuitive. It's flat out wrong. The intuition that production is lower or will be lower because of the disaster is correct. What is measured when activity picks up is NOT increased production over what would have been produced. It is production that replaces what would have been produced. And it is production that could have made more available to individuals but for the disaster.

What economists should be saying is that one will see increased production but one shouldn't see this as necessarily "good" but rather see it as the cost of the disaster.

Comments for this entry have been closed
Return to top