On defining "recession"

The precise definition of "recession" seems to be topical in Canada right now. (I know this because my daughter phoned to ask me the definition.) It's a mug's game. I won't play. (Do geographers waste their time arguing about the precise definition of "mountain"?)

I hear that the "technical" definition of "recession" is two consecutive quarters of (real) GDP growth below zero. But what the hell is the meaning of "technical" in this context? I only learned that "technical" definition by reading newspapers, not by reading economists. If I want the "technical" definition of "horsepower" I will ask an automotive engineer, and I figure I will get a more precise and accurate answer that will be useful if I want to compare two cars' engines. Well, the so-called "technical" definition of recession is certainly precise, but it's neither accurate nor useful.

What's so special about two quarters? What's so special about zero? What's so special about GDP? Precisely nothing.

There is absolutely nothing "technical" about the "technical" definition of recession. It's an arbitrary definition of recession, not a technical definition of recession.

If an agricultural economy produces less and less for 6 months of the year, every year, after peak harvest, would we say it goes into recession every winter?

If an economy with a steadily shrinking population has declining GDP year after year, would we say it is always in recession?

Well, we could say that, if we wanted to, but would it be useful?

GDP fluctuates over time. Macroeconomists try to explain why it fluctuates over time. We gain absolutely nothing by adopting some arbitrary cutoff to convert a continuous variable into a binary variable {recession; not in recession}. All we are doing is throwing away information. Precision is stupid. It's like defining "bald" as having less than 7 hairs on your head. Perfectly precise, and perfectly useless.

Which does not mean that "recession" is a meaningless concept, any more than "bald" is a meaningless concept.

Try asking a biologist for the precise definition of "species". The existence of ligers does not mean that lions and tigers are all the same. It's fuzzy, but not meaningless.

There's nothing special about zero, and there's nothing special about GDP either. GDP is just one way of measuring economic activity (and the definition of GDP is pretty arbitrary too, if you look at all the nitty gritty details of what gets included and what gets excluded, and how goods are valued). It would make exactly as much sense to have a "technical" definition of recession as two consecutive quarters of negative growth in total hours worked. Or two consecutive quarters of a rising unemployment rate. Or rising unsold inventories. Or whatever.

People notice patterns. And when people notice patterns, they want to try to explain those patterns. And the explanation of those patterns, and the definition of what is and is not part of the pattern, tend to go together. Long before there were GDP data, people had noticed a pattern in economic activity. They called that pattern the trade cycle, or the business cycle. Sometimes trade was faster, and it was easier to sell things; and sometimes trade was slower, and it was harder to sell things. The "conjuncture" is an alternative name for the trade cycle (more common in other languages than in English), and it's a good name, because it suggests a pattern of different variables all fluctuating together. To pick just one of those variables in the pattern (like GDP) is a mistake. To pick just one of those variables in the pattern, and choose some arbitrary cutoff like zero growth as the dividing line between recession and no-recession is an even bigger mistake. It's a pattern of co-movements, dammit, not a binary variable!

"Recession" is a theoretical concept. How you define it depends on your theory of what causes it. Because a useful definition according to one theory would be a less useful definition according to another theory.

For example, I see recessions as caused by an excess demand for the medium of exchange (money). An excess demand for money will make it harder to sell goods (including labour, and assets) for money, which causes a reduction in the volume of trade of goods for money. According to my preferred theory of recessions, barter trade will actually increase in a recession, so I would exclude barter trade from my definition. And any fall in production of goods is only a fall relative to what would have happened without an excess demand for money, and not necessarily a fall relative to the previous quarter, and is only a side-effect of the increased difficulty of selling goods for money. A fall in GDP, if it happens, is a possible symptom of recession, and not part of the recession itself. And a fall in GDP could be caused by something other than a recession. Like winter, for example. And GDP falls like a stone for two days every weekend, but I wouldn't call that a "recession".

A real business cycle theorist, who thinks that recessions are caused by technology shocks, and have nothing to do with money, would think my distinction between monetary and barter trade not a useful distinction when it comes to defining recessions. We aren't going to define "recession" the same way. The pattern he's seeing is not the same pattern I see.

But there's an election on, and I know there's going to be a lot of arguments about whether Canada is in "recession", and whether the government is responsible. ("No" and "no it's the price of oil" are my (tentative) answers so far). And that precisely stupid inaccurate useless arbitrary "technical" definition of recession is going to be getting a lot of play.

[Update: for journalists in particular. Let's stop calling the two quarter negative GDP growth thing the "technical" definition of a recession. Let's call it the "quick and dirty" definition of a recession instead. At least that won't mislead people into thinking it's authoritative. And it's our economists' fault for not having explained this to you a long time ago. But in our defence, it isn't something that's super easy to explain. And we do tend to get a bit tongue-tied around journalists.]

[Update 2: TLDR version: hearing 2 successive quacks is neither necessary nor sufficient for saying it's a duck.]

Comments

[Sandwichman]

A rant I can agree with almost wholeheartedly.

[JKH]

Agree with the points about arbitrariness, and the trenchant uselessness of a debate about whether we're in one or not.

But the work of NBER is not technical? More generally, there is such a thing as technical analysis of financial markets. It's a generally accepted term.

I'm also reminded of your aversion to the use of the term "money market" in the way that it is generally defined. But its a term that's been around at least as long as you or I have been breathing. I understand your objection to it, but it exists.

I'm also reminded of the absolute asininity of discussions about saving that are replete with arbitrary home grown, competing definitions of saving. There's one that serves all purposes just fine.

[Nick Rowe]

Sandwichman; that must be a first!

JKH: The NBER group look at a lot of things, not just the "2 quarters negative GDP growth" thing. The CD Howe has a Canadian version, including our own Stephen Gordon. I would certainly prefer whatever they come up with to the "2 quarters" thing, but I still can't really see the point, unless it's simply to try to replace the 2 quarters thing with a judgement call that is a bit less arbitrary and focused on one variable.

(If I set up a corporation, that buys a new car, and leases it back to me, that's "saving", according to the standard definition.)

[Hugo André]

Of course this measure is arbitrary but that doesn't mean it's useless. The definition works fairly well because it captures most of the economic episodes that economists generally consider to be recessions while ignoring smaller bumps in the road. If GDP growth were as seasonal as the agricultural economy you're describing then obviously we would need a different definition (as would a country with large seasonal variations in GDP).

Of course it does nothing to help economists in their work but journalism is famously bad with subtlety (you decide whether to blame the journalist or the reader for this) and having a simple dividing line like the "technical definition" means they get it right most of the time. What would be the alternative? Either you have a "you know it when you see it" approach which gives journalists plenty of room to exaggerate any minor fluctuation in output or there are economists trying to explain how complex it is and how hard it is to define an economic slowdown.

[Nick Rowe]

Hugo: fair point. Just wish we didn't call it the "technical" definition though. Because that makes it sound like the equivalent of the SAE (or whatever authoritative source) definition of "horsepower". It's the exact opposite of authoritative/expert/theoretically sound, which is what "technical" conjures up in my brain. Bunch of people in white lab coats and black-rimmed specs giving the matter some serious thought, rather than something grabbed off the shelf for a quickie. "Quick and dirty definition" would be a lot less misleading.

[JP Koning]

"We gain absolutely nothing by adopting some arbitrary cutoff to convert a continuous variable into a binary variable {recession; not in recession}. All we are doing is throwing away information. Precision is stupid. It's like defining "bald" as having less than 7 hairs on your head. Perfectly precise, and perfectly useless."

I'd argue the same about the arbitrary cutoff between money and non-money and your distinction between barter trade and monetary trade. ;)

[Stephen Gordon]

I prefer the word 'conjuncture' as well - it's the label for my file folder with all my current analysis stuff.

[Scott Sumner]

Nick, Good post, a few comments:

1. In the US, the technical definition of recession is NOT two consecutive quarters of falling RGDP. The technical definition is whatever the NBER calls a recession.

2. In 2001 the US clearly had a recession without two consecutive quarters of falling RGDP.

3. In the US (at least since unemployment data began1950) there are no borderline recessions. Each business cycle is either clearly a recession, where everyone agrees it's a recession, or clearly not a recession. That's an important stylized fact that (as far as I know) is not explained by any theory. It's as weird as if there were only earthquakes above 7 or below 4 on the Richter scale, but nothing in between. Unemployment either rises by more than 2.2% or by less than 0.8%, never anything in between. I don't think this lack of mini-recessions occurs in other economies--it's just the US.

https://research.stlouisfed.org/fred2/series/UNRATE

[Nick Rowe]

Hugo: and we can't blame journalists for thinking the "technical" definition is technical, unless we economists take the time to explain to them that it isn't. Which is one of the reasons I wrote this blog post.

JP: as Wittgenstein said: just because there is no sharp cutoff between "bald" and "hirsute" doesn't mean there's no distinction. Same with species. Or mountains and valleys. There are clines, but reality is lumpy. Rainbows have a pattern, even though there's no sharp cutoff between blue and green.

Stephen: Ah, but you work in French, which gets this one right!

When I go on the CD Howe page for the Business Cycle Council, there seems to be a missing link to the methodology section. All I can get is an Excell spreadsheet which lists a dozen indicators. Am I doing it wrong?

[Nick Rowe]

Scott: Thanks. I would agree that "What the NBER decides to call a recession" is a lot more truly technical that "two quarters negative GDP growth". But I can't count the number of times I have read that the 2 quarters thing is a "technical recession" up here.

Your point 3 is very important. It's part of the pattern, and needs to be explained. What is happening in Canada right now seems to be more of a borderline case. (Though, before Simon van Norden comes on, I would say I don't really trust the GDP data. And employment seems to be holding up reasonably well so far.)

[rsj]

There is a sense that the economy is bi-modal -- it's either growing or shrinking, with a phase transition in between. You can ask -- is this water a liquid or is it a gas, and the question makes sense, even though there is a period of time when water is boiling off or condensing.

This video is really cool:

https://www.youtube.com/watch?v=-b3PF89TLOM

[Nick Rowe]

rsj: see Scott's point 3 above. That to me is an interesting part of the pattern. We also notice that the downslopes usually tend to be steeper than the upslopes (relative to trend), so the recession phase is a little more sharply (or less fuzzily) defined than the recovery phase.

[Hugo André]

Nick, that I can definitely agree with. I've got nothing much to add except I don't think many journalists would want to use a "quick and dirty" definition. It just sounds too... dirty.

Might the lack of mini-recessions be an indication that Leijonhufvuds corridor hypothesis has something to it?

[john]

if there is a notion of a 'trend'-whatever that means to different people-then deviations below that trend can be seen as 'recessions' but what happens when there's no trend?

also, time in economics is a bit shaky as a category isn't it? I mean, how long is the 'short term' or 'long term' measured in days, months etc

Time is a qualified concept and so is a 'recession' so what matters is who is doing the qualifying part of the definition

Also, when you're saying that money's in short supply you mean that the equilibrium interest rate is too low?

thanks!

[Pete Bias]

Feel the same way about growth. To me, growth is society having more stuff. So if we recover from a (poorly defined, of course) deep recession at a 2% clip, we're not growing, we're recovering. Big difference. But "official" GDP rates of growth just show the 2%; no asterisk.
Oh well.

[Nick Rowe]

john: I definitely agree that we should look at deviations from trend when trying to identify the trade cycle. A big enough slowdown in growth, for an economy that normally grows quickly (like China), even if it never goes negative, would still be *one* symptom of recession.

"..but what happens when there's no trend?" Not quite sure what you mean by this. Maybe:

1. The trend is a flatline. OK, then we are going to look for the same deviations below that flatline.

2. It's a random walk with drift, so shocks are permanent. Maybe look at deviation from past drift? (I will leave that one to the econometricians who try (desperately) to identify unit roots.)

"Also, when you're saying that money's in short supply you mean that the equilibrium interest rate is too low?"

NO. ABSOLUTELY NOT! Sorry, but would take me too far off-topic to explain.

[Linda]

Commenting on a macro post is 'way outside my usual activity, but....a long (long) time ago, I did learn that a recession was defined as two quarters of negative growth, and I have no problem calling this a "technical" definition. Aren't all definitions technical, and arbitrary? Whether a "recession" by this definition is now, or ever was, meaningful, is a completely different issue, maybe akin to the difference between connotation and denotation?

[Kailer]

You need to mail this to every journalist in the country. Testify!

[Nick Rowe]

Linda: can you remember where you learned that 2 quarters definition? I remember I was at Western, in grad skool, in the late 70's or very early 80's. And I know it wasn't from any of my macro profs, or any macro textbook. Because I remember being surprised to read it in some newspaper or magazine. "Oh, is *that* how recession is defined? Says who?". Because it didn't match up with anything we were doing in macro class.

And I was just reading the Wikipedia page on "horsepower", and it really does get "technical" in the sense of being defined by bodies of engineers under various tightly-defined tests and conditions.

[Tom Brown]

Sorry, I can't resist: I don't know about "arguing" but there's this:

"The volume begins with an introduction to how mountains are defined..."

Interesting that one author is from Minnesota State University... he must be a "mountain theorist."

Plus there was that whole to-do about Pluto and planets. OK, I'll stop.

[Egmont Kakarot-Handtke]

Worthwhile Canadian filibuster?
Comment on ‘On defining "recession"’

Let us call the economist's bad habit to nudge any question to the point of inconclusiveness, overload, nausea, or inconsequential pluralism of manifest contradictions the Hicks drive.

“As far as I can make out, there are relevant and important senses in which all these statements are each of them right and each of them wrong.” (Hicks, 1939, p. 184)

To recall, real science is about true/false and not about multiple senses or anything-goes. Economists as would-be scientists are known to be sloppy: “The truth is, most persons, not excepting professional economists, are satisfied with very hazy notions.” (Fisher, quoted in Mirowski, 1995, p. 86)

This goes some way to explain the failure of economics. “I think it is the lack of quite sharply defined concepts that the main difficulty lies, and not in any intrinsic difference between the fields of economics and other sciences.” (von Neumann, quoted in Mirowski, 2002, p. 146 fn. 49)

Of course, in most everyday situations there is not much use to quarrel over exact definitions. It is a bit different in the scientific context. In economics in particular, haziness has two tangible merits. First: “By having a vague theory it is possible to get either result.” (Feynman, 1992, p. 159) and second: “Another thing I must point out is that you cannot prove a vague theory wrong.” (Feynman, 1992, p. 158)

Because it brings one on the safe side, vagueness, inconclusiveness, wish-wash are all elements of a conscious/unconscious survival strategy. “With enough fog emitted, almost anything becomes possible.” (Mirowski, 2013, p. 344)

This is why all half-wits, dilettantes, and shell game players applaud the silly Post Keynesian slogan: “It is better to be roughly right than precisely wrong” while scientists subscribe to “It is best to be precisely right than roughly wrong.”

In economics, habitual intellectual sloppiness had disastrous consequences. Neither orthodox nor heterodox economists understand the two most important phenomena in the economic universe: profit and income (2014; 2014). This is like medieval physics before the elementary concepts force and mass were clearly defined and understood.

There seems to be complete ignorance among economists that they have nothing to offer in the way of a scientifically founded advice.

“In order to tell the politicians and practitioners something about causes and best means, the economist needs the true theory or else he has not much more to offer than educated common sense or his personal opinion.” (Stigum, 1991, p. 30)

Of opinion there has been always more than enough in economics. Because they cannot tell the difference between profit and income neither the proponents of the Walrasian nor of the Keynesian approach have the true theory that could help to fix a crisis or recession. In view of humongous intellectual messines, whether economists can define recession makes no difference at all.

Egmont Kakarot-Handtke

References
Feynman, R. P. (1992). The Character of Physical Law. London: Penguin.
Hicks, J. R. (1939). Value and Capital. Oxford: Clarendon Press, 2nd edition.
Kakarot-Handtke, E. (2014a). Economics for Economists. SSRN Working Paper Series, 2517242: 1–29. URL http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2517242.
Kakarot-Handtke, E. (2014b). The Three Fatal Mistakes of Yesterday Economics: Profit, I=S, Employment. SSRN Working Paper Series, 2489792: 1–13. URL http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2489792.
Mirowski, P. (1995). More Heat than Light. Cambridge: Cambridge University Press.
Mirowski, P. (2002). Machine Dreams. Cambridge: Cambridge University Press.
Mirowski, P. (2013). Never Let a Serious Crisis Go to Waste. London, New York, NY: Verso.
Stigum, B. P. (1991). Toward a Formal Science of Economics: The Axiomatic Method in Economics and Econometrics. Cambridge, MA: MIT Press.

[john]

I guess I was thinking something along the lines of your point 2: there is ambiguity about the nature of the trend (stochastic vs deterministic etc) and also, anything that can be said about the presence or absence of a 'trend' is an ex-post exercise on the data-in that sense when the actual 'deviation' occurs, a) you don't know what the actual trend is in order to call it a 'deviation' in the first place and b)you get to call it a 'deviation' after some time when the system might be in a totally different place altogether

Obviously you can have estimates of the trend based on extrapolating on past data, now-casting and all that jazz but again that is an econometric exercise- it's not the actual trend.

I understand also that a part of your argument is that a 'recession' is not *only* about the dip in gdp and I think I can understand your point. But then again, if you do have a stylized model of the economy, then some of the variables will tend to to be pro-cyclical and others counter-cyclical. In that sense, tracking gdp is analogous to tracking 'all those other' variables that are positively correlated with it, in the *model*. In that sense, if you have a model and (some) things move together in it you get 1) a way to talk about trends and deviations and b) to focus on a few indexes characteristic of the state of your model of the economy

of course, like you say, there are other models with different properties, emphasizing different aspects

the thing is that, given the state of the economy (I assume there is one independent of the models), either there is one model that is correct and all the others are false-and it gets to call a recession a 'recession'-or there are more than one models of the economy correct and then what happens? Let's assume they can be further separated into those that are more general and somehow encompass some less general ones. Again, let's assume we end up with just 2 general models. Which one is the 'correct' one? Isn't it the case that either one of them is the 'correct' model corresponding to the 'true' state of the economy or that the state of the economy is itself dependent on the model we're using to describe it and in that sense, everything, including recessions, is relative to the definitions in the model?

anyway, thanks!

[Andrew_FL]

We can't define recession, but Potter Stewart er...Wesley Clair Mitchell knew one when he saw it.

[Steve Ambler]

Nick - I think the C.D. Howe is trying to fill the role of an NBER North to become a quasi-official arbiter for what qualifies as a recession. Scott's point (1) will apply: a recession will be whatever the Business Cycle Council decides to call a recession. You're right that the link is bad. It's supposed to refer to the Commentary piece written by Philippe Bergevin and Philip Cross. Here's a working link: [Link Here NR]. Interestingly, the paper actually ranks Canadian recessions on a scale: there seems to be more gradation in Canadian recessions than in U.S. recessions. The first half of 2015 is shaping up to be interesting. Don't know if you paid attention to the June trade numbers released Wednesday. For what it's worth, given that merchandise exports and imports each have a weight of about a quarter of GDP, a 4.8% increase in exports (by volume) and a 0.9% decrease in imports gives a (4.8+0.9)x0.25 = 1.425% boost to June GDP (that number comes out at the end of the month). That by itself wipes out the cumulative 0.8% decline in GDP from January to May and the approximately 0.3% decline in the first two months of this quarter. It will take a relatively steep decline in domestic demand to give us even a "quick and dirty" recession.

[nivedita]

I don't understand the controversy here. The exact boundary between a "slowdown" and a "recession" is clearly not going to matter much, but this is just like saying the exact boundary between a "tropical storm" and a "hurricane" is not going to matter much. Hurricanes still have a precise sustained wind speed cutoff of 64 knots. We don't think any less of the NOAA or meteorologists for that.

And I disagree that what you call a recession can depend on your model. That sounds completely bogus -- recessions are real phenomena independent of your model, just as storms are real phenomena independent of weather modeling. Different models may explain this phenomenon in different ways, but they don't get to pick what a recession is.

[Livio Di Matteo]

Well here is my two cents worth. When teaching intro economics, I've simply referred to the two consecutive negative quarters of growth as a "traditional" definition of a recession. It is a convention that has arisen over time and provides a simple definition-which quite frankly is what 98 percent of people are looking for.

[Nick Rowe]

Thanks for the link Steve.

One bit in the Bergevin-Cross paper caught my eye:

"The notion that a recession is defined by two or
more consecutive quarterly declines in GDP has
become well entrenched in popular discussions.
The origins of the consecutive-declines guideline
go back to a mistaken interpretation of a simple
statistical observation by the NBER that, in
practice, recessions in the United States lasted at
least six months (Moore 1967). Lay people, anxious
to penetrate the byzantine process used at the
time to assess cycles, quickly jumped on this as a
rule even though it was just a statistical artifact.
Indeed, the NBER itself has never used consecutive
quarterly declines in GDP as a definition of a
recession. A good example is the dating of the onset
of the recent recession in the United States: when
the NBER announced that the US economy had
entered recession at the start of 2008, consecutive
declines in total GDP had not occurred.6"

What a laugh! So the quick and dirty definition was a sheer accident.

[Nick Rowe]

nive: "Different models may explain this phenomenon in different ways, but they don't get to pick what a recession is."

Suppose I had a theory that recessions are caused by exactly the same thing that causes output and employment to decline every weekend (people decide to take a coordinated rest). Wouldn't it make sense for me to say that the Canadian economy goes into recession every weekend? It seems silly to have two different names for the same thing.

Suppose I have a theory that the morning star is the same thing as the evening star. Wouldn't it make sense for me to call them both Venus?

Try explaining the difference between "weight" and "mass" to a pre-Newtonian.

Facts are funny things. Yes, we can't pretend the world is anything we want it to be, but the facts we see depend on our theories.

[Chris J]

Nick, thank you. I now know /exactly/ what I sound like when I explain why Pluto isn't a planet.

[Frances Woolley]

Nick, great post! F.

[Egmont Kakarot-Handtke]

The Humpty Dumpty methodology
Comment on ‘On defining "recession"’

With regard to definitions the representative economist easily takes side with Humpty Dumpty. "’When I use a word,’ Humpty Dumpty said in rather a scornful tone, ‘it means just what I choose it to mean -- neither more nor less.’ ‘The question is,’ said Alice, ‘whether you can make words mean so many different things.’ ‘The question is,’ said Humpty Dumpty, ‘which is to be master — that's all’.” (Carroll, Through the Looking-Glass)

That is not how science works. The freedom or arbitrariness of definition is a methodological illusion. It applies only to the first definition. Subsequently, one has to make sure that every new definition is consistent with the preceding ones. Overall consistency cannot be achieved in the economist's cavalier fashion: “The only way to arrive at coherent languages is to set up axiomatic systems implicitly defining the basic concepts.” (Schmiechen, 2009, p. 344)

The fact of the matter is that economists got the fundamental concepts income and profit wrong. Keynes is a case in point. The formal core of the General Theory is given with: “Income = value of output = consumption + investment. Saving = income - consumption. Therefore saving = investment.” (Keynes, 1973, p. 63)

This is rather elementary mathematics and it should not be too hard to get it right. Actually, the fault in Keynes's two-liner is in the premise income = value of output. This equality holds only in the limiting case of zero profit in both the consumption and investment good industry. Profit does not appear in Keynes's elementary formalism. That is, he in effect talks about capitalism without profit. The simple reason for this analytical blunder is that Keynes never got the profit theory right. This sad fate is shared by the representative economist (2014) -- among many others by Nick Rowe.

The correct relationship is given with this equation:
https://commons.wikimedia.org/wiki/File:AXEC09.png

It says: the business sector's monetary profit Qm is equal to distributed profit Yd plus investment expenditure I minus the household sector's monetary saving Sm (2014, Sec. 3). Alternatively: The business sector's retained profit Qm-Yd is equal to the difference between investment and saving.

In short, household sector saving is never equal to business sector investment, that is, all I=S models including IS-LM are false.*

From this follows in turn that Nick Rowe's blog post of 2011** is logically defective as well as the recent posts on the Origin of Specious blog.***

The problem here as everywhere is the false definition of profit which in turn leads to a false definition of income which in turn leads to a false definition of saving. All mistakes together produce a closed and stable framework of self-delusion. Post Keynesians are trapped in this logical fallacy since more than 70 years.

Egmont Kakarot-Handtke

References
Kakarot-Handtke, E. (2014). The Three Fatal Mistakes of Yesterday Economics: Profit, I=S, Employment. SSRN Working Paper Series, 2489792: 1–13. URL
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2489792.
Keynes, J. M. (1973). The General Theory of Employment Interest and Money. The Collected Writings of John Maynard Keynes Vol. VII. London, Basingstoke: Macmillan.
Schmiechen, M. (2009). Newton’s Principia and Related ‘Principles’ Revisited, volume 1. Norderstedt: Books on Demand, 2nd edition. URL
http://books.google.de/books?id=3bIkAQAAQBAJ&printsec=frontcover&hl=de&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false.

* See also the cross-references http://axecorg.blogspot.de/2015/01/is-cross-references.
html
** http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/08/is.html
*** https://originofspecious.wordpress.com/2015/08/05/2420/

[Bob]

Sorry to intrude on this but can I ask you a question:
"yep, if you pay the interest on those bonds by printing even more money, as opposed to raising taxes."
Nick, I want to come back to this, is this still your view on MMT and forms your mental view on it, inflation worries etc.
MMT just says bond interest is just a type of welfare payment to certain individual with "rights" like child benefit or disability payments.
Any other Magical Properties to Bond Interest are propaganda on those maintaining the corporate welfare through.
Pensions can be managed by payment of a Universal Pension to all.
Interest payments are spent and generate tax revenue. Spending= Income
This is then spent etc. It generates an amount of tax and saving, each time, every time.
The government may then "borrow" the savings.
The government will get ALL of its spending back as tax unless someone saves some money down the line.
See:
http://www.3spoken.co.uk/2014/04/taxation-government-investment-each.html?m=1
What is your view on this article?

[Bob]

As to recession, I would argue it is entirely as hoc. The reason is you have recession is some industries at the same time as others.
IMV, The govt should try to aim for low unemployment rate and target this.

[notsneaky]

A fall in output caused by a fall in aggregate demand relative to output that would have been produced had demand remained the same.

Still some theoretical constructs but I think pretty much captures it. And yes, this means we have had many fewer recessions than we think (I would call a fall in output due to a supply shock, like oil prices or whatever, a "fluctuation", but not a recession)

[nivedita]

@Nick, Take weather models. They don't model hurricanes per se. They model a certain class of storm systems, a subset of which are fairly arbitrarily labeled "hurricanes". What they actually model are things like wind speed, spatial extent of storm, mean velocity of storm etc. In the same way, an economic model does not model recessions. It models variables like output, consumption, employment etc. Certain values for these variables or their rates of change are fairly arbitrarily labeled "recessions".

While these labels are to some extent arbitrary, they are not completely arbitrary. It would not make sense to call every gust of wind that exceeds 64 knots a hurricane. A definition of hurricane that works on Earth may not be appropriate on Jupiter. In the same way it would not make sense to call a drop in GDP for a couple of days a recession, and the definition that works for the United States may not be appropriate for, I don't know, Nigeria.

Further, if you had a model that explains recessions by the same thing that causes output to drop on weekends, just at a vastly larger scale, that would be fine, and quite interesting, just as the same model that describes a fluid flow inside a rotating bowl might explain hurricanes at a different scale. Nobody would call the system inside the bowl a hurricane, though.

If you had a model that says the morning star and the evening star are both Venus, that would be a prediction of your model -- and your model would be vacuous if you simply defined them both at the outset to be Venus: the whole point of the model is that two apparently different phenomena are really the same object. So I'm not sure what you're getting at here -- you don't call the morning/evening star Venus, you call Venus the morning/evening star.

Your other example of weight vs mass is more interesting -- and yes, if you come up with a model that says the phenomenon we label as "recession" is really three fundamentally different things that produce the same observable effects, because we are currently missing what makes them different, then it may make sense to give them three different labels, assuming your model is correct and becomes widely accepted as correct. So in physics, while we distinguish between mass and weight post Newton, and mass and rest mass post Einstein, consider instead inertial mass vs gravitational mass. To the best of our knowledge, these two are identical, so aside from the people who design experiments and theories to resolve whether they are truly identical or merely very close, the rest of physics still has only one concept for mass.

[Nick Rowe]

nive: "and yes, if you come up with a model that says the phenomenon we label as "recession" is really three fundamentally different things that produce the same observable effects, because we are currently missing what makes them different, then it may make sense to give them three different labels, assuming your model is correct and becomes widely accepted as correct."

That's sort of roughly how some of us macroeconomists think, only it's two different things, not three. There's "true" recessions, caused by drops in aggregate demand; and there's those declines in GDP caused by drops in aggregate supply, that look a bit like "true" recessions at first glance, but if you look more closely they are different in important ways, so we should call them a different name. But RBC theorists disagree, and say all recessions are caused by AS, and AD doesn't matter, so we should give them all the same name.

[Nick Rowe]

In 1937, for example, Haberler, in Prosperity and Depression started out by saying (something like) IIRC: 'Some drops in GDP are caused by things like bad harvests, but that's not what we mean by "recession"'.

[NoniMausa]

To further confuse the topic, it seems to me that a recession is a delta function -- that is, some measure has changed, become lower in relation to an earlier baseline. As such, we cannot absolutely say a recession is bad or growth is good. It all depends on the baseline that is optimal for the country or society in question.

I spent a year in Alberta at a time when the province was booming, with a great influx of workers, and oil money splashing all over. As a result the housing market was nuts. For six months I couldn't find an apartment in Vegreville, an hour outside Edmonton, and had to live in a friend's spare room. Why? Because rentals were occupied by the families of oil workers who themselves were living and working in Fort Mac. I suppose by many measures the economy was doing fine, but not for the ordinary citizens.

So, recession itself isn't a useful term. But is there an alternate measure of overall wellbeing? It seems that much of political skill lies in picking a useful metaphor. How about an air-mattress measure? If the Canadian economy were an air mattress, I'd say we were at about 80% inflated at the moment. Mostly comfortable, but if you tried a hasty change of position you could bruise an elbow or wobble right off your bed. We still beat the US, where many people are lucky to get a length of bubble wrap, but we could do better.

[John Goldin]

I'm not a geographer, but indeed there has been a need to construct quantitative (and arbitrary) definitions of mountains. See the definition section of the Wikipedia article on "mountains". The plot of the film The Englishman Who Went Up a Hill But Came Down a Mountain (1995) revolves around the definition of mountain (and lampoons the arbitrariness).

[Kevin Donoghue]

A Snowdonia peak has been downgraded to the status of a hill.

"Guidelines state a peak needs to be 2,000ft (610m) high and have the 49ft (15m) height difference in order to be classified as a mountain."

http://www.bbc.com/news/uk-wales-north-west-wales-33833974

[am]

Definitions in economics have to be international. As much as they are not consistent internationally there will be no meaningful comparisons between the economic performance of countries.
Definitions of recessions are internationally accepted as two consecutive quarters of negative gdp growth.
Definitions can be changed. But this will require rewriting of textbooks, explanation of the new definition by the media to the public, etc. This will bring in the question who is doing this and why. Politicians and politics will be the answer by the public.

[Donald A. Coffin]

FWIW, here's what the NBER has to say:

http://www.nber.org/cycles/recessions.html

[Jim Groves]

goodonya from a country (Oz) that has been in recession but because it isn't "technical" we ignore it!