Refuted economic doctrines #8: the superiority of flexible labor markets

by John Quiggin on May 22, 2009

According to the latest Eurostat data*, the unemployment rate in the US was equal to that in the EU-15 in March, and is now likely to be higher. Writing in the NY Times, Floyd Norris refers to the conventional wisdom that flexibility inherent in the American system — it is easier to both hire and fire workers than in many European countries implies that unemployment should be lower (at any given point in the business cycle) in the US than in Europe.

Although this is the conventional wisdom, the research on which it was based (by Lazear and others) has long since been qualified or refuted. I looked at this in the context of the Australian debate about unfair dismissal laws a few years back. Although the early research supported the simple view that more flexibility = more jobs, later research yielded the conclusion that employment protection laws lower the variance of employment and unemployment but have no clear effect on the average levels.

When you look at the institutions involved, this is unsurprising. Although Norris refers to restrictions on hiring and firing, the reality is that restrictions on hiring are nowhere very important. Even in the extreme case of Spain, they amount to a few weeks of salary. The big differences are in costs of firing, and it is the anticipation of these costs that acts to constrain hiring in periods of expansion.

Advocates of the US system make much of the deterrent to hiring associated with employment protection laws, but they ignore the other side of the coin. When the economy is contract, employment protection laws do in fact protect employment (if they did not, they would have no adverse effect on hiring either).

On this basis there is nothing surprising in what we are seeing. EU unemployment rates should be higher in expansions and lower in contractions, which is exactly what is required for lower variance.

Which is better? The big problem with strong employment protection is that it tends to create an insider class of well-protected permanent employees and an outsider class who are either unemployed or assigned to some form of semi-permanent “temporary” status (see, for example, the tenure system in universities). But, comparing the EU and US as a whole, the opposite seems to be the case – there is a sharper class divide, and less social mobility in the US than in the EU.

More on this from the Center for Economic Policy Research (h/t Brad DeLong).

  • In response to various comments, this is harmonised data adjusted to remove the effects of differences in national definitions. The choice of definition still makes a difference – for example, the harmonised definition refers to the civilian non-institutional population. If prisoners were treated as unemployed, this would raise the US rate by about 1.5 percentage points and the EU rate by about 0.2 percentage points.

{ 88 comments }

1

Selfreferencing 05.22.09 at 11:36 pm

John, I’m a pretty big fan of your work on expected utility theory, but I find that as a blogger, you seem to be a different man, with wildly different standards of evidence. As a scientist, you know that a single data point is compatible with an infinite number of theories and cannot by itself refute any theory decisively. So isn’t it ludicrous to argue, as you have for the last year or so, that the current crisis refutes any particular economic doctrine? How could a single crisis prove to be a decisive refutation of anything at all?

And isn’t it remarkably fortuitous that the crisis just so happens to refuse all the economic doctrines you already rejected?

2

Selfreferencing 05.22.09 at 11:40 pm

err … ‘refute’, not ‘refuse’.

3

Righteous Bubba 05.22.09 at 11:41 pm

How could a single crisis prove to be a decisive refutation of anything at all?

This seems to me to be crazy talk, but I am not an economist.

4

John Quiggin 05.22.09 at 11:56 pm

“And isn’t it remarkably fortuitous that the crisis just so happens to refuse all the economic doctrines you already rejected?”

Not exactly true, as I discuss here. I was broadly sympathetic to the New Keynesian program before the crisis, much less so now.

As regards decisive events, I disagree. Given the difficulties with conclusive econometric tests, much of our capacity to test economic theories depends on how they perform in times of economic crisis.

5

OneEyedMan 05.23.09 at 12:12 am

Bryan Caplan wanted to make the following bet with the CEPR report (http://www.cepr.net/documents/publications/US-EU-UR-2009-05.pdf) , it sounds like Dr. Quiggin might be interested in taking the bet since the Authors would not:

“The average European unemployment rate for 2009-2018 (i.e., the next decade) will be at least 1% higher than U.S. unemployment rate. The bet will be resolved when Eurostat releases its final numbers for 2018. I’m happy to bet each of the three authors $100 at even odds. Will they accept? P.S. By 1% I of course meant 1 percentage-point.”

6

steven 05.23.09 at 12:17 am

As a scientist, you know that a single data point is compatible with an infinite number of theories and cannot by itself refute any theory decisively.

I am not a scientist, but this is flagrantly bullshit.

7

Barbar 05.23.09 at 12:38 am

If you know that a number is greater than 100, that information is compatible with an infinite number of theories (“125″,”5 million”, “1 trillion”,…) and cannot by itself refute any theory (“25”) decisively.

8

notsneaky 05.23.09 at 1:46 am

Hmm back when I actually cared to blog I did a post on this:
http://notsneaky.blogspot.com/2007/09/bit-more-on-effects-of-job-protection.html
the post was prompted by some very good comments here on CT by derrida derider.

9

bert 05.23.09 at 2:16 am

“… comparing the EU and US as a whole, the opposite seems to be the case …”

How about what we’ve seen recently in France: the self-protective strikes by public sector and other favoured workers, in parallel with the far more self-destructive riots by the “racaille” in the banlieues. The first group are very much an insider class, and boast a string of prime ministerial scalps. The second are marginalised socially and geographically, with youth unemployment rates at or above one in four.

I’d argue that French labour laws show up in the figures whatever point of the cycle you’re at. Why else would you consistently find France at the top end of the productivity tables? It’s not because they’re worker ants – they’ll proudly tell you that themselves. It’s because French capitialism results in a mix of factors of production that’s comparatively light on labour.

Whether these points generalise to the European Social Model as a whole I don’t know. Perhaps any assertion that there’s more social mobility in the US than in the EU is a previously-refuted doctrine, and I missed the memo. But if you look at the CEPR graph, there’s a whole economic cycle (roughly corresponding to the Clinton years) where the US figure is consistently around four points better than Europe’s. Just in terms of social outcomes, that ain’t nothing.

Unpleasant to be on the Mark Steyn side of any argument, but there it is.

10

John Quiggin 05.23.09 at 2:32 am

“Perhaps any assertion that there’s more social mobility in the US than in the EU is a previously-refuted doctrine, and I missed the memo.”

You have indeed missed the memo. This doctrine was refuted long before the current crisis, and so hasn’t been included in the series. Rather than select sources, I suggest using Google.

http://www.google.com.au/search?q=US+%22social+mobility%22+europe&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a

11

Selfreferencing 05.23.09 at 2:43 am

John:

Ok, that’s nice to know about the alteration. It still seems fishy, but I appreciate the reply. I take it that this is only a 5% change or so, but that is better than I can say for most people, maybe I’m included.

As for econometric modeling, I’m extremely skeptical of the ability of the incredible complexity of the market to be broken down in such a way that it can refute a particular economic theory. Of course, it could be used to interpret the crisis as evidence in tension with particular views, but that is a far cry from a refutation. Maybe my sympathies with the late Hayek’s complexity theory are too strong here, but I would like to see the reasoning. Do link me to the argument, if you can.

Steven: Just read a philosophy of science reader, man. Look up ‘Quine’ and ‘holism’ and ‘underdetermination’ in the index.

12

Keith M Ellis 05.23.09 at 3:48 am

There’s other meanings of “flexibility” than those you’re addressing here. Don’t you think that geographical labor flexibility probably is superior? As a non-economist, I always feel that geographical labor flexibility should be given importance more in line with that given to the flow of trade. Am I confused about this?

13

Keith M Ellis 05.23.09 at 4:02 am

Selfreferencing, steven is right that the statement is bullshit, your obfuscating condescending citations notwithstanding. A single data point can, indeed, refute a theory decisively.

At any rate, anyone who manages to namedrop recursion, Hayek, Quine, holism, complexity, and undeterminism in a single comment doesn’t deserve to be taken seriously. You’re trying way too fucking hard and more bring to mind that guy in the bar who won’t shut up than, say, an esteemed intellectual who convincingly wins an argument.

14

a 05.23.09 at 4:49 am

“As a scientist, you know that a single data point is compatible with an infinite number of theories and cannot by itself refute any theory decisively.”

If economics is a science, then its theories should be falsifiable by a single data point. So is your point that economics is not a science?

15

Finnsense 05.23.09 at 6:17 am

If we’re talking about labour markets then rather than compare the US and EU, why can’t we compare nations inside the EU that have radically different labour markets? Denmark moved to a very flexible system and at one point had unemployment below 2%. Even now it’s significantly below that in the US and EU – and that of Finland and Sweden that otherwise have similar systems. I think something has to be explained there.

As a postscript I’m always a bit wary of unempoyment figures. We know the US and EU calculate them differently and some of us probably remember the McKinsey study that said Sweden’s unemployment rate was actually nearer 16% than the 5% that was reported – which in turn lead to an analysis by Dean Baker et al showing the same was true for the US. So who to believe?

16

John Quiggin 05.23.09 at 7:09 am

Finnsense, you can also look at Employment/Population ratios. For men, they tell pretty much the same story as the unemployment data. For women, they vary a lot depending on social institutions that affect participation rates – higher than the US in Scandinavia, but a lot lower in Germany, IIRC.

Keith, the US almost certainly benefits from more geographical mobility, but that only makes the point in the post more striking.

As a general point, doctrines in economics tend to rise and fall in a dialectical fashion. So when I say “refuted”, I don’t really mean “proved false now and forever’. I mean “much less plausible than they seemed pre-crisis, and in need of some radical revision”.

17

musical mountaineer 05.23.09 at 7:13 am

Perhaps it’s just the measure of my ignorance, but I find this counterintuitive. Where job protection is greater, shouldn’t people tend to stay in the same job longer, and thus be less socially mobile?

I read a couple of the links from the Google search you kindly provided above. Social mobility is defined most often as the likelihood of moving in one’s lifetime from the bottom to the top quintile. That sounds nice, but it obviously must produce losers as well as winners. If one person goes from the bottom quintile to the top, another must fall from the top to the bottom, or four others must descend by one quintile, or something like that. By definition, you can never reduce the population of the bottom quintile.

It occurs to me that social mobility will increase if you just import a quintile’s worth of Arabs whom nobody will employ, and stuff them in the banlieues. Voila! Everybody who was previously in the bottom quintile has now moved up. I’m not saying that’s what’s happening, because I don’t know. But it shows that high social mobility doesn’t necessarily mean good things are happening.

Likewise, low social mobiliy is perfectly compatible with everybody getting stinking rich. The rising tide lifts all boats, and all that. Again, I don’t claim to know what’s actually going on, but perhaps social mobility is not the best measure of people’s life opportunities.

18

notsneaky 05.23.09 at 7:44 am

“Where job protection is greater, shouldn’t people tend to stay in the same job longer, and thus be less socially mobile?”

Intra vs. Inter job social mobility. And usually the job protection feeds back into this.

“Social mobility is defined most often as the likelihood of moving in one’s lifetime from the bottom to the top quintile. That sounds nice, but it obviously must produce losers as well as winners.”

Hmmm, yea there’s a lot of really bad social mobility measures out there, which basically survive based on the fact that they accord with someone’s political ideology somewhere. But social mobility is supposed to measure the amount of “churning” in the distribution – so you can have the same distribution at two different points in time but it’s different people (or families) at different places in it (in fancy talk this means that your measure of social welfare does not satisfy the condition of “anonymity”). But there’s some simple ones which make a lot of sense – for example the correlation between father’s and son’s income, or, if you like, between a father’s and son’s rank in the distribution (as of now, the data on mother’s income is still subject to problems). As it turns out both of these are positive (richer fathers have richer sons, fathers higher up in the distributions have sons higher up in the distribution) but perhaps as not much as some people think (IIRC corr coff about .35 or so, with wide variance in the lit)(As a complete aside Greg Clark argues that social mobility was helluva higher back in the feudal days, at least for England, with a corr coff something like .05).

Basically it’s learning by doing vs. flexibility.

19

StevenAttewell 05.23.09 at 8:01 am

Question: How does the long-term picture of U.S vs. E.U unemployment look if one compensates for the different ways in which the U.S and E.U measures unemployment? I know I’ve read material that’s suggested that the U.S and Swedish unemployment rates are actually pretty similar if one takes the U6 measure as opposed to U3.

20

R.Mutt 05.23.09 at 9:30 am

21

Barry 05.23.09 at 9:32 am

a 05.23.09 at 4:49 am

“If economics is a science, then its theories should be falsifiable by a single data point. So is your point that economics is not a science?”

Selfreferencing 05.23.09 at 2:43 am

” As for econometric modeling, I’m extremely skeptical of the ability of the incredible complexity of the market to be broken down in such a way that it can refute a particular economic theory. “

a, the answer is ‘yes’. Selfreferencing does not believe that economics is a science, in the sense of producing falsifiable theories. Note that SR stated a strong view of the nonfalsifiable belief, rather than the view that this particular theory can/can not be refuted with the data at hand.

22

Down and Out of Sài Gòn 05.23.09 at 9:56 am

As a scientist, you know that a single data point is compatible with an infinite number of theories and cannot by itself refute any theory decisively.

But it’s quite easy for a single data point to refute an infinite number of theorems in mathematics, if the theorems are sufficiently stupid. For every given integer n >= 0, let ‘Silly Theorem n‘ be the proposition “There are no integers less than n” (which can be expressed as a well formed formula). Choose -1 as your single data point counterexample.

23

bert 05.23.09 at 10:04 am

John, at #10 you pick the one part of my comment where I’m explicitly not making an assertion, and you choose that assertion to refute. The rest you ignore.
(That said, no hard feelings – I could have phrased it clearer.
I’m nowhere near Mark Steyn either, who I understand tends to witter on about demography, so while I’m about it I should perhaps make that clear too.)

Your alternate take on the French case would be interesting if you care to give it.

24

David Weman 05.23.09 at 10:08 am

“Question: How does the long-term picture of U.S vs. E.U unemployment look if one compensates for the different ways in which the U.S and E.U measures unemployment?”

No one seem to’ve tried to calculate it. The US excludes people that Eurostat counts. Using the eurostat definition should give you consistently higher numbers. But how much higher?

I imagine if the data had supported pro-corporate policies instead, someone would have done the work.

25

Hidari 05.23.09 at 10:15 am

‘As a scientist, you know that a single data point is compatible with an infinite number of theories and cannot by itself refute any theory decisively’

I mean absolutely. Anyone who claims that one so called ‘flight’ by the Wright brothers refutes the impossibility of mannned flight doesn’t understand that this phenomenon could be explained by an infinite number of theories. For example, pixies could have come and lifted the Wright brothers’ so-called ‘aeroplane’ off the ground. Or else maybe the Earth just fell down in space about thirty metres and the ‘aeroplane’ stayed in the same place?

And don’t get me started on the moon landings.

26

belle le triste 05.23.09 at 10:32 am

what — in a materialist context* — does an “infinite” number of theories actually mean?

the argument seems to be this:

i: a single datapoint is compatible with an infinite number of theories
ii: infinity is the same as everything evah
iii: hence a single datapoint is compatible with all possible theories
iv: hence cannot by itself decisively refute any theory

sadly for this logic, a single datapoint is also INcompatible with an infinite number of theories, and thus (by ii-iv) decisively refutes all possible theories also

*assuming this is synonymous with “as a scientist”

27

andrew_m 05.23.09 at 11:10 am

(still off thread):

belle le triste,

I’m not sure whether I can explain what having an infinite number of theories “actually means”, but the concept gets used routinely when making inferences from data.

When someone make a maximum-likelihood estimate using statistics, it’s a two- step process:

1. Consider the set of hypotheses (i.e. small theories) that the quantity being measured takes each of an infinite number of possible values.

2. Select the hypothesis that’s least improbable given the data, and take that as the estimate.

28

Henri Vieuxtemps 05.23.09 at 11:47 am

I dunno about the significance of costs of firing. Capitalism tends to work around these things, there are always loopholes. Anecdotally (from what my Italian friends tell me), the level of permanent employment goes down, temporary contracts up.

It’s complicated; for example, apparently they have some government-paid furlough program: you can cut working hours for your employees (say, to a three-day week), and the government will pay the balance. Sort of like getting unemployment benefits while still being employed. How do you account for something like that?

I suspect what probably matters more is a large public sector; it’s stable and perhaps even countercyclical.

29

musical mountaineer 05.23.09 at 12:08 pm

“…there’s a lot of really bad social mobility measures out there, which basically survive based on the fact that they accord with someone’s political ideology somewhere.”

Bingo! I was about to say, social mobility studies say more about the researchers’ perceptions of class envy/struggle than they do about individuals’ chances to live a prosperous and fulfilling life. A better question might be what percentage of people end up better off in absolute terms than their parents.

Another thing to consider, since these studies are about comparing different societies’ policies and outcomes, is the extent to which societies are entangled with one another. For instance, suppose that a more flexible labor market somehow leads to greater innovation and entrepreneurship; other kinder and gentler societies would benefit from the resulting technological discoveries. Now that’s a pretty vague and untestable hypothesis, but it shows the principle that each system’s outcomes depend, to some extent, on the others’, and it may not be wise for anyone to say “they should be more like us”.

30

Keith M Ellis 05.23.09 at 12:19 pm

Personally, I think it’s more likely that he somehow thinks that if a single data point is compatible with an infinite number of theories, therefore a single data point can disprove no theories. Because, as belle suggests, somehow “infinite number of theories” is being mistaken for “all theories”?

It can be quite challenging to puzzle out someone’s apparent irrationality.

31

Robert 05.23.09 at 12:38 pm

Perhaps economists ought to pay more attention to economists who were correct all along. My name links to a paper by Graham White, one souch Australian economist.

32

Matt 05.23.09 at 12:38 pm

When the economy is contract, employment protection laws do in fact protect employment
I don’t doubt this, but want to look a bit further before I decided what to make of it. Does this lead to more firms failing, as they are locked into contracts they can’t get out of? Does it lead to more firms needing government support? Does it lead to firms not expanding when they would like to because of uncertainty? I suspect there are at least some answers to these questions, but I think we need to know these answers before we feel very good about the conclusions either way.

The big problem with strong employment protection is that it tends to create an insider class of well-protected permanent employees and an outsider class who are either unemployed or assigned to some form of semi-permanent “temporary” status
Again, before we know how to judge this result, I’d want to know a lot about who it is that ends up in this status. If it’s like getting a tenure track job in the university or not, then it’s tragic and unpleasant for those who don’t get the job but not, all things considered, a very serious social problem, as the “losers” can usually move to something else fairly easily and the decisions, while of course involving luck and connections, are somewhat merit-based and not inherently tied to immutable characteristics. But if the group assigned to outsider status happens, quite consistently, to be dark and “foreign”, then we might wonder both whether the good gained is worth it and whether it’s maintained for reasons that are not legitimate. (Of course the less-mobile groups in the US also have traits that might make us suspect that it’s something other than pure merit or luck that determines who moves and who doesn’t, too, but my point is just that we need to see if the “semi-permanent group” is of a particular sort to determine how bad this problem is, and that I suspect that it’s a bigger and deeper problem in much of Europe than the raw numbers suggest.)

33

Willem van Oranje 05.23.09 at 1:09 pm

musical mountaineer:
Bingo! I was about to say, social mobility studies say more about the researchers’ perceptions of class envy/struggle than they do about individuals’ chances to live a prosperous and fulfilling life. A better question might be what percentage of people end up better off in absolute terms than their parents.

Nice one. You don’t like the results of the studies so you trot out the old “it must be class envy”.
Google the studies and you’ll find that they specifically tried to answer your “better question”: what percentage of people end up better off in absolute terms than their parents. Not only that. They also tried to answer the question: what percentage of people ended up WORSE off in absolute terms than their parents.

34

Willem van Oranje 05.23.09 at 1:24 pm

Knowing that the EFCA is a hot topic right now in some circles in the US, I would like to point out that several European countries with low unemployment nrs have also strong unions with collective bargaining for entire industries.

I think that would be lovely material to see some heads explode :-)

35

Barry 05.23.09 at 2:16 pm

Willem van Oranje 05.23.09 at 1:24 pm

” Knowing that the EFCA is a hot topic right now in some circles in the US, I would like to point out that several European countries with low unemployment nrs have also strong unions with collective bargaining for entire industries.”

But they don’t count because they’re all small, or backwards, or whitehomogeneous, or….

36

mart 05.23.09 at 2:58 pm

As regards unemployment rates, some time ago over here (the UK), IIRC the government changed a lot of people who were unemployed onto “sickness benefit” which meant they weren’t included in the regular measures for unemployment = made government look good. I don’t know if the US or other EU members have done something similar, but I think here there’s been a difference of several percentage points between official and actual unemployment rates as a result.

37

Willem van Oranje 05.23.09 at 3:13 pm

Mart, the data used for the graph are from Eurostat and are harmonized.
For instance, Eurostat says that unemployment in the Netherlands is 2.8 while according to the Central Bureau of Statistics in the Netherlands itself, unemployment in NL for the last three months is 4.6

38

tired of blogs 05.23.09 at 4:01 pm

I agree with your general message, but…and I am not an economist…I always took the argument that flexible labor markets are superior to mean not that unemployment will be lower at ANY point in the business cycle. Rather, unemployment may rise much more sharply during a crisis, thus possibly passing the levels of unemployment observed in countries that make it harder to fire people (if relative flexibility doesn’t mean that people may get fired at a faster rate, what does it mean? and if the rate’s faster, the levels may converge or flip). But it should recover much more quickly and return ultimately to lower levels.

Now, again, I don’t personally buy that this is always true, but this seems to me to be the argument about flexible labor markets implicit in, say, the general approach of the Economist. Particular economists may have made more ambitious claims, and perhaps those are generally accepted by economists, but I’m not aware of it. Do tell!

39

tired of blogs 05.23.09 at 4:10 pm

belle le triste —

Visually, the idea that a single point is compatible with an infinite number of theories might be explained like this:

Statisticians make scatterplots of the relationships between variables. There’s a dot for each observed outcome with its x-coordinate equal to the value of the causal variable and its y-coordinate equal to the value of the effect variable. If you’re trying to explain presidential election outcomes, you might use recent changes in the economy as the cause and the vote share of the incumbent as the effect. You could have dozens of dots on that scatterplot, and the basic statistical technique would be to draw a line that captures the relationship between the dots. In the example, if the economy improves, presumably the data will show that the incumbent’s vote share also improves, and the line will be positive.

Geometrically, two points form a line. So if you have only two observations, they’ll define the line. If you have more than that, statistics provides various ways of generating an “average” line if the data don’t fall perfectly onto one.

But if you only have one point, you can’t even define a line, just as a matter of simple geometry. The relationship has to produce the effect value you see at that cause value — in other words, the line has to pass through that point — but any line that passes through that point satisfies the condition, whether positively sloped, negatively sloped, flat, etc. Could be a curve, even. You just can’t tell from one dot.

40

steven 05.23.09 at 4:29 pm

Yes, you can draw infinitely many curves through a given point. There are also infinitely many curves that do not pass through a given point. (I’d be curious to know if these two infinities are of different cardinalities, btw.)

41

armando 05.23.09 at 4:53 pm

No, those infinities are the same. Just an eensy weensy continuum.

42

Keith M Ellis 05.23.09 at 4:59 pm

steven, they must be, mustn’t they? Just an intuition (the same as yours, I suspect).

tired of blogs, I don’t think belle, or anyone else for that matter, had any strong objections to the claim that a single data point supports an infinite number of theories. That’s not the objection. The objection is his assertion that a single data point cannot disprove any theory. Belle attempts to scry a chain of thought that supports this.

Speaking of intuition, rumbling deep in my intuitive netheregions is the suspicion that a carefully formulated theory might, in fact, only have a finite number of data points, or even only one, which support it—which would make both of Selfreferencing’s assertions of this nature untrue. Nevertheless, that a single data point supports an infinite number of theories does seem to be inarguably true, in general, and is unobjectionable.

43

Robert Waldmann 05.23.09 at 4:59 pm

I have tenure and I love it. I absolutely detest employment at will which makes employees dependent on the arbitrary will of employers.

However, one swallow does not make a summer. The case against the US labor market is that, when bankers totally destroy the economy, US unemployment is as high as European unemployment for the first time in IIRC 27 years.

Even the flexibility fanatics agree that easy firing causes lower unemployment in the worst of times. That is, make predictions which fit the facts exactly.

I’d say the case for employment protection is based on noting that Europe is not one economy and that some European countries manage consistently low unemployment while giving workers some rights to keep their jobs.

I’d look to Holland and Austria not the EU-15 for the case for legal protection of job security.

44

mpowell 05.23.09 at 5:42 pm

Okay, maybe I’m missing something, but why is it a fundamental tenet of this theorem that a flexible economy must give better unemployment numbers at every stage of the economic cycle? Wouldn’t it be significant to say that flexible markets give better employment over the course of the entire cycle? Also, I think it’s really dubious to move this theory into the category of ‘refuted theories’ based on this one example, even if you want to stick to the requirement that employment be better at every stage of the cycle. The recession is hitting some countries harder than others for reasons entirely unrelated to their employment flexibility. In such a case, the ‘refuted theory’ claim is too strong, I think.

45

Kenny Easwaran 05.23.09 at 9:08 pm

The point about the single data point is that we know that in almost all observations, there’s some margin for error. If we have a lot of data points, we can hopefully see some confirmation of an underlying trend, together with getting an estimate for what sorts of things are relevant to the error, and what the size of the error term might be. If we just have one data point, then we don’t even get an estimate on the size of the error term, so any theory can be saved by appeal to auxiliary hypotheses. (i.e., someone can look at the current crisis and say that their favorite unemployment theory remains unfalsified because of some unusual factor at work, like government support of the auto industry, or unusual tariff policies, or something else).

As for the actual topic of the discussion, I thought it was particularly interesting seeing that Ireland and Spain, the two European countries that got most caught up in the bubble over the past few years, are the only ones in the EU15 that seem to have had any sort of meaningful increase in unemployment numbers over the past three years, and their increases were way above that of the US.

46

Z 05.23.09 at 9:19 pm

Wouldn’t it be significant to say that flexible markets give better employment over the course of the entire cycle?

It would be significant, though significantly weaker than the statement that flexible job markets always outperform less flexible ones, and this has been the position held by a non negligible part of the political class in this last 10 years or so (at least in my country). Switching from “always outperform” to “outperform on average” is very significant in terms of public policies because: after all, it is not unreasonable for a society to choose an “insurance policy” against catastrophic crisis events “paid” by a higher average rate of unemployment to avoid catastrophic crisis events. That said, from the little I know about this subject, and according to the direct links of this very post, even the much weaker formulation is empirically dubious.

None of this diminishes the importance of the insider/outsider problem, which has I think a drastic impact on my society.

47

musical mountaineer 05.23.09 at 9:30 pm

“Google the studies and you’ll find that they specifically tried to answer your “better question”: what percentage of people end up better off in absolute terms than their parents. Not only that. They also tried to answer the question: what percentage of people ended up WORSE off in absolute terms than their parents.”

I’m sure some of the studies do address the question, Willem. And some of them don’t. The ones that don’t tend to assume that social mobility is an absolute good, and tend to be used as sticks to beat up on us free-market types. I hope I sufficiently disclaimed expertise in my first post, but today isn’t the first time I’ve poked around the internet on this very issue. With regard to at least some of what passes for research out there, my objections are cogent and ought not to be hastily dismissed. As for those studies of trends in absolute wealth, my guess (only a guess, mind!) is they don’t make such good talking-points for advocates of government control.

Now, I had actually written a paragraph on why my “better question” might not be so illuminating, but deleted it before posting. I think we can take it as read, that most people in most systems for most of the last 50 years have ended up better off than their parents, thanks to the progress of culture and technology. So the answer is rather dull, apart from the exceptions to the trend: the downwardly-mobile, who must have suffered disasters or squandered the family fortune. And, well, that’s social mobility.

48

Righteous Bubba 05.23.09 at 9:53 pm

The point about the single data point is that we know that in almost all observations, there’s some margin for error.

This ship is unsinkable! I don’t think the “single data point” quite covers the multiple catastrophes that have happened around the world, or the degree of exposure some countries had vs. others.

49

musical mountaineer 05.23.09 at 9:53 pm

Robert Waldmann wrote: “I have tenure and I love it. I absolutely detest employment at will which makes employees dependent on the arbitrary will of employers.”

I’m old enough to have had a midlife crisis, and I’ve been working since I was a kid. My longest-running job was five years, and the next runner-up would be not more than two years. I’ve worked in restaurants, in manufacturing and prototyping, in mechanical engineering, in software, and in emergency medicine. I have always been subject to the will of employers, but I only got fired once, and I had another job before I could clean out my desk.

I love it. For me, switching jobs has always been by far the fastest track to higher earnings and greater satisfaction (which is why I get puzzled when people say that less-flexible labor markets make greater social mobility). It would have severely cramped my style to have to deal with a load of government oversight every time.

I have no objection to others having tenure. But I would argue, at least, that all labor markets should not be the same. Different people have different ideas of fun.

50

Jordan DeLange 05.23.09 at 10:02 pm

Selfreferencing is confusing two related, but somewhat distinct points, neither of which really establishes his claim. When SR says “a single data point is compatible with an infinite number of theories ” SR is presumably referring to the underdetermination of theory by data. And it is pretty plausible that there are many distinct theories compatible with any finite data set. But this, as many commentators have already pointed out, does nothing to establish that a theory cannot be refuted by a finite data set, or even a single data point.

To get that “[A single data point] cannot by itself refute any theory decisively” SR is presumably (at least, given his hand waving towards Quinean holism) referring to the Quine-Duhem thesis. This supplements the point about underdetermination by also maintaining that any theory and any observation can be made compatible by suitably revising other auxiliary hypotheses or theories. The idea is something like “hey, I used my microscope and saw X, but the theory says Y, so either the theory is refuted, or my background theory of how my microscope works is wrong.” And since we always appeal to some other hypotheses or claims when making observations and drawing implications of those observations for our theories, the claim is that any observation itself can be made to be compatible with any theory.

However, this is a radical position, and even if you were to accept it, it wouldn’t follow that the economic crises could not refute any economic theory. As long as we don’t revise any background assumptions, a conflict between the data provided by the crises and particular economic theories can easily serve to refute those economic theories. And this is certainly the case if we understand “refute” as John Quiggin suggests, as something like render much less plausible. Unless SR identifies what particular auxiliary hypotheses he thinks should be revised to render the data of the economic crises compatible with the economic theories Quiggin identifies, SR is simply obfuscating the issue at hand by throwing out at best barely relevant philosophy of science.

51

John Quiggin 05.23.09 at 10:28 pm

Bert @ #23. Sorry if I you felt I gave your question short shrift – presented with a rhetorical question (did I miss the memo?), I felt it reasonable to respond with a straight answer (Yes, indeed).

I don’t know enough about France to comment in any detail on the riots and similar events, though IIRC there was some good discussion from other CTers. My general point is that although the insider-outsider labour market problem is a serious one for Europe, the statistical evidence is clear that Europe, as a whole, has higher social mobility than the US. Given how poorly this fact is understood in the US, the use of impressionistic data points that appear to support the opposite view (a la Mark Steyn, as you point out) is highly misleading.

52

Omega Centauri 05.23.09 at 10:57 pm

I’d like to look at “flexible labour market” via a couple of thought experiments.

The first is from the standpoint of a firm and its employees during a crisis, such as today. In the flexible case, layoffs, or involuntary reductions of hours/wages will come about faster. In the shortest timeframe, this hurts the workers. A little later in the cycle, does the firm with the inflexible market survive? In the flexible case, moderate harm is certain, in the inflexible case the results tend to bifurcate, either everyone comes through OK (keeping their jobs), or everyone loses theirs as the firm fails.

The second thought experiment is job exchange. Suppose two workers doing different jobs have initially been miss assigned: Bob would be much more productive and happy with Steve’s job, and likewise Steve would excell at Bob’s job. In a flexible workplace, should this fact be discovered, the exchange could take place, and productivity would improve. In this case, no change in employment level ocurrs (at least due to direct effects), but average productivity is affected.

If the main proposed social good from imposing greater inflexibility is income security, then wouldn’t a more robust safety net, be a more economically efficient way to acheive it? Of course Europe has both a more robust safety net, and less flexible labour market. I guess the corellation of policies overrides the substitutability of policies.

53

Cranky Observer 05.23.09 at 11:24 pm

> A little later in the cycle, does the firm with the inflexible market
> survive? In the flexible case, moderate harm is certain, in the inflexible
> case the results tend to bifurcate, either everyone comes through
> OK (keeping their jobs), or everyone loses theirs as the firm fails.

You are leaving Wall Street out of the equation. Since 1995 Wall Street (and the banking/investment community in general as affecting non-public corporations) have increasingly imposed the Cisco Rule(tm) on an ever-growing percentage of US firms: “50% growth/year and 40% gross margins _forever_ or you will be destroyed”. I can’t count the number of good and fundamentally sound entities I have seen destroyed by these Wall Street-imposed “stretch goals”. In the so-called ‘flexible’ case the 40% gross margins continue, taken out of the workers’ hides. In the so-called ‘inflexible’ case it isn’t so clear what happens, but there is at least some chance that the workers’ lives will be saved by reducing the grossly unrealistic gross margin targets. Sharing the pain with the overclass as it were.

Cranky

54

bert 05.23.09 at 11:30 pm

I once worked with a guy who got miss assigned. Everyone was very broad minded, but the whole thing was weird I have to say.

John, we cool.
One last point: there’s been several suggestions in the thread that there’s a measurement problem with comparing employment figures. I’m sure that’s true, although I’ve not looked in any detail. But my sense is that while there may be a problem, it’s not an explanation. If that’s too complacent I’m sure someone will put me right.

55

b9n10t 05.24.09 at 1:03 am

Cranky’s point is a good place for this anecdote:

My bro in law was, well, not exasperated exactly, but struck that his health insurance company was laying off personal this fall. Because of the economic crisis, Wall Street simply expected labor force reductions, even in firms that were making decent profits, had a decent market share and a sound business model.

I’ve a feeling that labor market inflexibility (against firing, in the Euro. mode as Quiggin mentions) may work to secure long term investment (in human capital) that is especially important in an increasingly highly trained, specialized work force. May labor market inflexibility be a check on the ravages of short term profit-maximization?

56

b9n10t 05.24.09 at 1:04 am

personal = personnel.

57

StevenAttewell 05.24.09 at 2:36 am

David Weman:

How hard would it be to do the work of comparing the Eurostat numbers with U6 over time?

58

Omega Centauri 05.24.09 at 3:23 am

If I go back to my comment #52, I make an argument that an optimal policy might be high labour flexibility backed up by a highly robust unemployment safety net. Yet we seem to have two sorts of societies. Those that care about worker welfare, have both inflexible labour laws, and a robust backup system. Those that don’t put a premium on worker welfare, have neither. The supposedly optimal solution is not choosen. Why is that the case?

59

Omega Centauri 05.24.09 at 3:32 am

I’m having some difficulty understanding Cranky’s observations in #53. How do the Wall street mavens force bad business decesions upon managers? I can understand that their attitude might be able to temporarily cause their stock to go out of favor, but the effect should be limited -as wise investors would see great value in purchasing shares of a corp with only say 40% growth, and 30% gross margins at firesale prices. The effect of wall street’s greed should thus be greatly attenuated. Perhaps some managers were foolish enough to be bamboozled into such self destructive actions? I just can’t imagine it would be a widespread phenomena however.

60

Judd 05.24.09 at 4:58 am

There is no single safety net/job security regime in Europe. Some are similar to their neighbor’s systems. Some seem singular.

This is important as it is all too easy to claim for all of Europe the worst of the lot for whatever you wish not to have. Thus those opposed to a single payer system prattle on about the British Health care system and those who are fans of flexible labor markets love to site French labor laws.

61

StevenAttewell 05.24.09 at 5:33 am

Omega:

Hostile takeovers, shorting stock, lowering ratings, changing recommendations from buy/hold to sell, etc. Even wise investors tend to walk sideways if the market starts hammering the value of your stock.

62

magistra 05.24.09 at 6:39 am

Another issue about US unemployment statistics that’s probably relevant is the huge incarceration rates there. There’s a decent argument that a lot of the people who would be longterm unemployed in a European style system have instead been imprisoned in the US (which also conveniently acts as a makework programme for prison guards etc).

63

SJ 05.24.09 at 7:54 am

“I’d be curious to know if these two infinities are of different cardinalities, btw”

This question has already been answered, i.e., the cardinalities are the same, but I can explain the reasoning.

You have a point (x,y) in a plane. There is an infinite number of lines that pass through it, and you can identify them with a single real variable, say theta, the angle each makes with the x axis. The cardinality of a single real variable is aleph one.

Now, for each single one of those lines with angle theta, you can identify a different x or a different y such that the line no longer passes through the original point (x,y). So you need three real variables to identify all of the lines that do not pass through (x,y). The number of lines that don’t pass through (x,y) equals the number of points in 3-space, which still has cardinality aleph one.

64

JoB 05.24.09 at 8:40 am

I’m still waiting for the “refuted economic doctrines: economies of scale are linear with scale” and/or “economies of scale tend to benefit all economic agents”.

65

Hidari 05.24.09 at 9:23 am

‘I love it. For me, switching jobs has always been by far the fastest track to higher earnings and greater satisfaction (which is why I get puzzled when people say that less-flexible labor markets make greater social mobility). It would have severely cramped my style to have to deal with a load of government oversight every time.’

I’ve heard this argument before and it always strikes me as utterly bizarre. To begin with: ok fair enough. You don’t want tenure. But tenure and employment rights weren’t brought in to facilitate your ‘free and easy’ employment path. They were brought in so that bosses could hire and fire more easily. What YOU wanted was irrelevant.

But much more importantly…why do you think, with tenure, people lose the ability to quit their jobs? And what ‘government oversight’ do you have to face? Dunno where you live, but for most people outside North Korea, to quit your job you just hand in a letter to give three months notice. Or, if you’re not bothered about references, just walk out. What ‘government oversight’ is involved there?

66

steven 05.24.09 at 12:05 pm

Thanks, SJ!

67

Keith M Ellis 05.24.09 at 1:51 pm

“If I go back to my comment #52, I make an argument that an optimal policy might be high labour flexibility backed up by a highly robust unemployment safety net. Yet we seem to have two sorts of societies. Those that care about worker welfare, have both inflexible labour laws, and a robust backup system. Those that don’t put a premium on worker welfare, have neither. The supposedly optimal solution is not choosen. Why is that the case?”

Because the operative political opposition here is labor versus business, not whatever it is you have in mind (such as collective versus individual rights or market economics versus socialism). This is the fundamental insight required for making sense of American politics (and others, as implicit in your comment; but the US example is extreme). The GOP, especially, likes to cast all political conflict in terms of more versus less regulation, more versus less taxes, and other things closely associated with the supposed ideological divides in the US.

However, the reality is that the GOP’s true affiliation is to business, full stop. In the US, we have so-called “labor flexibility” along with a poor safety net—your least optimal solution—because the specific interests of business are “labor flexibility” and low taxation, which is incompatible with a robust safety net.

Now, in the wider domain, I think it’s true that, generally, the operative political opposition is labor versus business. This is an accident of history. Business had, and has, extreme power because of wealth. It has incentives to exploit labor. Politically, then, a natural counter to business becomes the creation and evolution of political labor. And that opposition defines the alignments of political parties around the world to this day. Labor wants, for obvious reasons, both “labor inflexibility” and a robust safety net; business similarly and in opposition wants both “labor flexibility” and a weak safety net. Both are sub-optimal. And so it goes.

68

musical mountaineer 05.24.09 at 3:25 pm

Hidari: “I’ve heard this argument before and it always strikes me as utterly bizarre.”

That’s because I didn’t state it accurately; I preferred to be concise. We’ll get to that in a moment.

“…tenure and employment rights…were brought in so that bosses could hire and fire more easily.”

Yes, well… What?

“…why do you think, with tenure, people lose the ability to quit their jobs? What ‘government oversight’ is involved there?”

None, in the sense that you mean. Also, I’m not exactly talking about tenure (which implies a probationary period with merit review etc.), but about employment rights generally. My point is, the phrase “less flexible labor market” means what it says.

Take France, where there is no such thing as “at will” employment. There, if an employer wants to fire someone, his first step is to hire a lawyer and prepare for battle. If the employee is not provably guilty of stealing or trying to burn down the shop, the employer may well lose his case and be stuck with the lazy slob forever.

Under these circumstances, employers obviously must exercise extreme diligence in hiring. They will hire only in the direst necessity. Because there is no scope in the law for probationary hiring, employers will more strongly prefer people they know in advance and can expect to get along with: friends and relatives.

So it’s easier to find the perfect woman than it is to get a job. And if you happen to have a job, you’ll be extremely reluctant to quit. The job market may be thought of as a pipeline or electrical circuit: the flow decreases no matter where the resistance is. So, just because I don’t literally have to talk to the Commissar if I want to quit my job, doesn’t mean that government oversight is irrelevant to my decision.

I understand most people prefer job security, and I won’t argue with most people. Starting a new job is stressful and arguably wasteful. But the French system must be miserable for everyone. Employers can’t get temporary help, or choose freely whom they will pay and spend their workdays with. Employees who happen to get a job they dislike face a steep climb to get a different job. Customers will regularly encounter dead-eyed slackers who don’t give a damn about service or quality of work, because nobody gets fired just for being surly and indolent.

One time, I narced out a co-worker (for being lazy, surly, and a mess-making meddler) and got him laid off. Now that is Not Nice, and my heart was uneasy about it. But then a bunch of my other co-workers expressed such joy and gratitude (not to me; they didn’t know I did it) that he was gone! He really was degrading the morale and productivity of the outfit. In France, he’d have remained there long after all the good people quit.

69

Laurent GUERBY 05.24.09 at 4:40 pm

Regarding comparison of OECD normalized unemployment number one should be very careful.

Tables on OECD stat site gives for 2007 (latest detailed data currently available on the web site) on male 25-54 the following data:

employment divided by population
France: 88.3% (hence 11.7% jobless)
USA: 87.5% (hence 12.5% jobless)

normalized unemployment rate
France: 6.3%
USA: 3.7%

So with a slightly higher relative working population in this age/sex group France has … a 70% higher unemployment rate than the USA.

This readily available data is unfortunately little known. I’m at a loss at why serious economist do compare “unemployment” numbers anymore, any idea?

70

Omega Centauri 05.24.09 at 5:57 pm

Keith @67, Thanks that was well stated. The less than rational outcome is a “natural” result of the political dynamic.

#68. It is clear that either extreme of the spectrum is seriously suboptimal. In the extreme flexibility case, workers will be reluctant to undertake any sort of longterm financial commitments, for obvious reasons. So even the employers will suffer fom a tougher economy.
Isn’t our problem that as a body politic we seem unable to see issues as a spectrum, and solutions as being a search for the optimal control point. Instead we gravitate towards ideologies that support one or another extreme. And special interest groups that want a locally optimal solution for their narrow niche os society at the expense of the whole actively support whatever ideologies are seen to help push things in the preferred direction.

71

Robert 05.24.09 at 7:34 pm

No theoretical basis exists for the proposition that more “flexible” labor markets lead to less unemployment. And, as far as I know, the data do not support this proposition either.

72

Peter Whiteford 05.24.09 at 8:40 pm

One of the recurring themes in this thread and many others on the internet is that the devious authorities have manipulated the definition of unemployment to minimise the unemployment statistics in one’s own country or in one of those identified as the best performers (so one’s own country isn’t really as bad as it looks).

In most cases this is simply not true, so long as one is careful to use the standardised ILO series that is available from sources such as the ILO or the OECD (as Laurent Guerby points out).

Yes, the ILO definition says someone is employed if they worked an hour a week in the survey week, but this is the case in all countries using the ILO definition, and most developed countries do this now.

So on this basic definition we are comparing like with like. It is true that the Thatcher government (and those in other countries) manipulated the unemployment statistics in their day to reduce the claimant count (the number of people receiving unemployment benefits), but this has no direct impact on the ILO measure.

It is also true that there are lots of Swedish workers off sick on any given day and who are counted as employed, but we don’t have comparable statistics for other countries to know how much of a difference this makes cross nationally. While it would be good to have an internationally comparable “at work employment rate”, sick people are not unemployed since they have an ongoing employment relationship and are paid. Nor are they disguised unemployed, since what this means is that Swedish employers are profitable enough to pay people who aren’t at work – this may be a work incentive problem, but it is not that Sweden is unable to generate these jobs.

It is also true that the US has a much higher proportion of its population in prison than any country in Europe, so if we assumed that all of the difference in imprisonment rates should be added to the US unemployment rate, it would be higher, but I think it can be calculated that it would never have added more than 1.5 percentage points to the unemployment rate.

So as Laurent Guerby pointed out stick to the OECD statistics particularly those for the employment to population rate – but unfortunately Laurent I think that you made an error in your specific example. The higher unemployment rate figures cover those less than 25 years of age and those aged 55 years and over where employment levels in France are well below those in the US, even though in the “prime age” group French employment levels are actually higher than in the US.

73

John Quiggin 05.24.09 at 9:12 pm

Following up on Laurent’s point, I’m aware of this point and have discussed it in the past, but the assumption
Unemployed=Population – Employed
is only a reasonable approximation for prime-aged males (and of course, not perfectly accurate even for them). So, rather than get into a fight about whether focusing on this group is reasonable, I thought I would stick to the harmonised unemployment figures.

Peter W, you say, correctly that adding in imprisonment would never have added more than 1.5 percentage points to the US unemployment rate. That’s true, but 1.5 percentage points is about half of the average difference in harmonised rates over the period since 1993.

74

bilejones 05.24.09 at 9:35 pm

Comparing these unemployment figures is meaningless. The BLS numbers have been politically bastardized for decades, for example a nasty little shit called Reich reduced the sample of inner city (read black) people by 20 % . I’m sure that the political filth in Europe have their own set of lies.
See John William’s Shadowstats.com

75

Martin Bento 05.25.09 at 4:16 am

When we speak of unemployment in terms of its ill effect on the population, shouldn’t we factor in that being unemployed is a much greater burden where the welfare state is less generous? I would like to see an “unemployment misery index” that accounted for this, by looking not just at the fact of unemployment, but its effect on standard of living.

76

Laurent GUERBY 05.25.09 at 8:38 am

#72, it’s not a “mistake” on my side, it’s more a “choice” :). When comparing country labour market you have a choice of indicators to use. Male 25-54 employment rate is my favourite for several reasons:
– accross time and cultures, this category always has to work
– leaves out end of studies and retirement, which are social choices, hard to say where the optimum is
– leaves out women, also useful for time based comparison (not the case here though, see below).
– alternative like unemployment are not objective measures since they are based on “effort” perception and not on a simple yes/no accounting observable like employment. And as shown by data unemployment even normalized is very dubious as shown by my examples.

#73, the issue is that the unemployment rate for this age/sex group should be about equal in France in the USA, but it show a 70% deviation! Doesn’t this show that unemployment measure is broken beyond repair? Why should it be wrong just for male 25-54 and not for other categories as well? I’m curious about your arguments or peer-reviewed papers on this topic.

As for women 25/54 in 2007 from OECD:
E/P
FR 76.1
USA 72.5

unemployment
FR 7.7
USA 3.8

So here we have clearly more women working in France (3.6 percentage point) relative to their population but unemployment is 102% higher in France! Even worse than for make 25-54.

Economists should get serious about the data they use at some point, don’t you think?

77

Ginger Yellow 05.25.09 at 8:53 am

“Perhaps some managers were foolish enough to be bamboozled into such self destructive actions? I just can’t imagine it would be a widespread phenomena however.”

Have you looked at the US newspaper industry? It’s all over the place. The Chicago Tribune Co, for instance, was making perfectly decent (but not huge by newspaper industry standard) profits until Zell bought them out with massive amounts of debt and began slashing costs. Now it’s bankrupt. And this isn’t a freak occurrence. Look at the NYT Co, Philadelphia Newspapers or the Rocky Mountain News

78

mpowell 05.25.09 at 10:51 am

77: Following up with this point, the idea that executives can afford to ignore the dictates of Wall Street is quite ill-informed. Some can get away with it, but it is not unreasonable to expect a stock value to change by as much as 50% downwards due to a negative Wall Street outlook. At that point, the P/E serves as a bottom but you are in serious buy out territory if outside investors see that the market attaches a negative valuation to your management team. As an executive, it is not just a problem with the board who are very interested in the share price, but also with outside vultures who will pursue higher margins. This has a lot to do with why I think the incorporation model used in the US deserves to be reexamined by liberals.

79

JoB 05.25.09 at 11:24 am

78: And another issue is that Wall Street compares across industry sectors forcing banks, food and car companies to converge on similar ratio’s, or else . In fact the mechanism of distributed ownership of increasingly big companies traded as if they all were in a primary business of them printing stocks is one of the largely non-discussed failures of the current system. The only real difference that Wall Street has allowed between companies is between established and new. The former have to generate cash by shedding employment and the latter are allowed to burn cash – up to the point they’re integrated in the former which is prior to the creation of substantial type employment. When ownership of companies is distributed and stock valuation an only measure to put all of them in front of the potential distributed owners, this must happen as far as I can see at least. A combination of issues: corporations that become too big, public ownership of the corporations via individual citizens de facto unfair to the citizens and concentration of a public power in non-elected officials (stock markets, Buffet’s, rating agencies, captains of industry and the odd politician elected to do something else than ensure the interests of private money).

80

Ginger Yellow 05.25.09 at 11:37 am

79: This is an issue I’m looking at quite closely when it comes to banks. The implication of the regulations being proposed at the moment is a much lower return on capital for high street banks. Now I think that’s a good thing – high street banks shouldn’t be chasing risk or leverage in a way that produces high returns on capital. But, given the way “Wall Street” works, it does suggest they’re going to have more difficulty raising capital in future than riskier investment banks, which may create some perverse incentives at the macro level.

81

JoB 05.25.09 at 11:56 am

80: yes, the issue is especially clear in the case of banks (the Dutch finance minister now has the courage to explicitly call for deconsolidating banks, by the way) but it is imho lots more generic than that. Anyway, as you point out: we’ll not be able to change the system for banks without an associated change in ‘Wall Street’ which is, maybe, why currently the attitude there is of denial: nothing really happened, governments overreacted and back to the order of the day (lay-offs, & optimizations & consolidations)

82

StevenAttewell 05.25.09 at 5:13 pm

mpowell: “This has a lot to do with why I think the incorporation model used in the US deserves to be reexamined by liberals.”

Interesting comment. Any suggestions?

Myself, I’ve always thought that Santa Clara could use some revisiting.

83

Ginger Yellow 05.25.09 at 5:34 pm

Steven, I don’t profess to have particularly well thought out opinions on the subject, but it’s a pretty mainstream idea (on the left, anyway), that directors of public companies should have explicit legal responsibilities broader than maximising shareholder value (which itself tends to be defined in pretty harmful ways). Basically, the principle is that society confers all sorts of benefits on oluntarily incorporated public companies s(including personhood, these days), so it should have social obligations in return.

84

Thomas Jørgensen 05.26.09 at 5:17 am

Honestly, measuring only male employment is absurd, because it renders any comparison between countries with differing degrees of gender equality and female labour market participation utterly meaningless. The labour of women does bloody well count. – Unemployment stats are not a very good measure to compare either, unless the countries involved measure this in the same way, which is rarely so. Employment as a share of total population is the only meaningful measure if you wish to compare the effect of diffrent ecomomic and social policies on labour market mobilisation, which is the entire point of this debate, no?

85

Laurent GUERBY 05.26.09 at 2:25 pm

#84 Thomas Jørgensen, I don’t think male 25-54 employment is affected much by the state of women employment, if it was given the massive rise of women E/P we’d see a widely different situation. As my data shows, the FR/USA discrepancy is even worse for women 25-54 than for male.

Also when doing time based comparison, like here:

http://fatknowledge.blogspot.com/2009/01/misleading-nature-of-unemployment.html

Male 25-54 is the only option otherwise you have to filter out in some way the rise of women E/P.

If you take employment/total population as quantity to maximise it means that you don’t want kids going to school and elders must work until they die. Not sure it’s a societal optimum :). That’s why 25-54 is good since it removes school and retirement from the picture and these categories don’t have an accepted optimal E/P whereas 25-54 optimal E/P is certainly close to 100%.

Of course you can play games and having measures redirecting jobs to 25-54 from 15-24 and 55+. Technically one can argue this is the case in France for 55+ with government sponsored layoff of elder employees, but with health care and other measures they usually don’t fall into unemployment or poverty.

I hope to have answered your concerns,

Sincerely,

Laurent

86

lemuel pitkin 05.26.09 at 3:08 pm

Again, though, the question is *why* we are making the comparison. There is no single right way to measure unemployment in the abstract, it depends on the context. Are we asking

– whether aggregate demand is sufficient to employ all available labor, i.e. how far we are from “full employment”?
– what fraction of the population are suffering the specific harms associated with involuntary unemployment?
– how much slack in the labor market is required to keep the threat of job loss effective and in general to maintain the power of capital over labor (the real objective of those who run our economies)?

There’s no one real unemployment rate — depending which of these questions (or some other) is being asked, the right answer will vary.

87

lemuel pitkin 05.26.09 at 3:09 pm

(I forgot, you can’t use dashes to indicate lists. Oh well, I guess it’s still readable enough.)

88

Gary 05.27.09 at 7:01 pm

I don’t see anything in this article that supports, let alone proves, you concluding assertion that “there is a sharper class divide, and less social mobility in the US than in the EU.” What justifies this conclusion?

Comments on this entry are closed.