Readers will, by now, have read the famous blog post by @sturdyalex in which it is claimed that, far from being lazy, Greeks are the hardest-working people in Europe. Veteran readers will know I have considered the factual basis of this argument here and here. However, not only has the piece gone viral since then, it's also resurfaced in this review on the New Statesman, reaching a whole new, and more opinionated, audience.
To forestall some of the strongest objections, I share Alex's disgust at the stereotypes and pseudo-anecdotes flying around and exploited for populist reasons abroad. I believe, as I've stated elsewhere, that the Greeks are intrinsically no better or worse, harder-working or lazier, than the Swedes, the British, the Germans or the Zimbabweans. I assure you that revisiting the data properly confirms this. Insofar as Alex's argument debunks the myths being peddled about us Greeks, I am fully supportive.
However, people's reading of Alex's data is becoming increasingly warped, and leaves some commentators wondering how such hard-working people are lagging so far behind in unit output. More to the point, it reinforces the wrongheaded view that Greece's labour market is not deeply dysfunctional and in desperate need of reform. Hence I need to revisit my original commentary and aim for a wider audience.
I will, furthermore, not rely on arguments such as 'yeah but look at productivity', which, while valid, essentially only serve to kick the ball into the long grass. I know that an hour spent sitting at a desk playing Spider Solitaire is not the same as an hour spent making stuff people will pay to buy, but I'm not sure one corresponds to Greeks and the other to Germans, and that's not our point right now.
No, my point is that the OECD data Alex relies on are deeply flawed and unsuited to international comparisons. If looked at in detail, they tell a very interesting story, just not the one some of Alex's readers would like.
You see the OECD figures do not compare like for like. They lump part-time and full time employees together without any adjustment so countries with a larger percentage of part-time employees will tend to have fewer average hours to show for it. As this OECD table will show you, part-time employment is very rare in Greece. To take the example most likely to raise eyebrows, 22% of German workers worked part-time in 2008, against 8% in Greece.The OECD also lumps together female and male employees, so any country which manages to get more women into the workforce also has fewer hours worked to show for it because women typically worker fewer hours. For background, only 56% of Greek working age women were economically active in 2008, against 71% of German working age women (source). Account for that, and for all the other ways in which the two labour forces differ, and, well, it gets complicated.
There are many simpler ways of getting the overall picture but the simplest is to compare like-with-like. A full-time male Greek employee to a full-time male German employee, and likewise for women and non-employees (proprietors, the self-employed, family members working for businesses, more or less anyone without an employment contract). I'll stick to German and Greek employees, partly because it's sensational to draw this comparison and partly because comparing Greece to all possible EU countries will make my tables way over-complicated.
I can draw this comparison in particular because the OECD data on European countries are more or less directly comparable as they are derived from the pan-European Labour Force Survey. As a passing glance at the OECD matadata will reveal, comparisons with non-European countries are actually not valid at all and the OECD discourages users from using the data in this way. The reason is that some countries use administrative or workplace-based data as opposed to the LFS' employee-reported data, which tends to underestimate hours worked (read here for a full review).
So, comparing full-time employees with a formal employment contract in 2008 (the reference year Alex uses), Greek men work only 2% more hours than their German counterparts, while Greek women work 1% less than German women (all figures in hours worked weekly below). Full-time non-employees in Greece, both male and female, work 3% fewer hours than their German counterparts. Part-time female employees in Greece only work 7% more hours than their German counterparts. (Data here, or see table below, which you can also download from here)
So that's the whole of the mainstream, more productive, labour market in a nutshell. The difference in hours worked on a weekly basis is nowhere near the staggering 48% implied by the OECD figures that Alex cites.
So what accounts for the difference in hours worked? For the most part it's the structure of the labour market. Fewer women and part-time workers in Greece, and a substantial fringe of workers that don't fit the traditional model of work.
The latter are very important. They are mostly part-time male employees, part-time non-employees of both sexes, and people with second jobs (data on those here). That's precisely the segment of the labour market that is shut out of fully-regulated work, not least by employment regulations. It's the kind of work, may I add, that our unions are, to this date, trying to regulate out of existence. Here, the difference is absolutely huge. A Greek man with a second employee job, for instance, works about twice as long as their German counterpart. Given the lower productivity in Greek workplaces, keeping working hours in mainstream jobs artificially low through employment regulation is not doing anyone any favours. A simple reading of Alex's data would suggest nothing's wrong. But there is plenty wrong.
There is much in this data that the European Left could use, if only they bothered to. The reading above is just one libertarian point of view. Getting to the bottom of why the Greek labour market includes a fringe working its metaphorical hairy knuckles to the bone is essential to understanding our plight. Unfortunately it doesn't play to the preferred narrative of evil neoliberals trying to talk the proud, untamed Greek worker down. Well the data aren't playing ball. What's it going to be, friends? Change the narrative, or manipulate the data? Do you want to win a flamewar, or do you want to understand the shit we're in? I'm not keeping my fingers crossed.
To forestall some of the strongest objections, I share Alex's disgust at the stereotypes and pseudo-anecdotes flying around and exploited for populist reasons abroad. I believe, as I've stated elsewhere, that the Greeks are intrinsically no better or worse, harder-working or lazier, than the Swedes, the British, the Germans or the Zimbabweans. I assure you that revisiting the data properly confirms this. Insofar as Alex's argument debunks the myths being peddled about us Greeks, I am fully supportive.
However, people's reading of Alex's data is becoming increasingly warped, and leaves some commentators wondering how such hard-working people are lagging so far behind in unit output. More to the point, it reinforces the wrongheaded view that Greece's labour market is not deeply dysfunctional and in desperate need of reform. Hence I need to revisit my original commentary and aim for a wider audience.
I will, furthermore, not rely on arguments such as 'yeah but look at productivity', which, while valid, essentially only serve to kick the ball into the long grass. I know that an hour spent sitting at a desk playing Spider Solitaire is not the same as an hour spent making stuff people will pay to buy, but I'm not sure one corresponds to Greeks and the other to Germans, and that's not our point right now.
No, my point is that the OECD data Alex relies on are deeply flawed and unsuited to international comparisons. If looked at in detail, they tell a very interesting story, just not the one some of Alex's readers would like.
You see the OECD figures do not compare like for like. They lump part-time and full time employees together without any adjustment so countries with a larger percentage of part-time employees will tend to have fewer average hours to show for it. As this OECD table will show you, part-time employment is very rare in Greece. To take the example most likely to raise eyebrows, 22% of German workers worked part-time in 2008, against 8% in Greece.The OECD also lumps together female and male employees, so any country which manages to get more women into the workforce also has fewer hours worked to show for it because women typically worker fewer hours. For background, only 56% of Greek working age women were economically active in 2008, against 71% of German working age women (source). Account for that, and for all the other ways in which the two labour forces differ, and, well, it gets complicated.
There are many simpler ways of getting the overall picture but the simplest is to compare like-with-like. A full-time male Greek employee to a full-time male German employee, and likewise for women and non-employees (proprietors, the self-employed, family members working for businesses, more or less anyone without an employment contract). I'll stick to German and Greek employees, partly because it's sensational to draw this comparison and partly because comparing Greece to all possible EU countries will make my tables way over-complicated.
I can draw this comparison in particular because the OECD data on European countries are more or less directly comparable as they are derived from the pan-European Labour Force Survey. As a passing glance at the OECD matadata will reveal, comparisons with non-European countries are actually not valid at all and the OECD discourages users from using the data in this way. The reason is that some countries use administrative or workplace-based data as opposed to the LFS' employee-reported data, which tends to underestimate hours worked (read here for a full review).
So, comparing full-time employees with a formal employment contract in 2008 (the reference year Alex uses), Greek men work only 2% more hours than their German counterparts, while Greek women work 1% less than German women (all figures in hours worked weekly below). Full-time non-employees in Greece, both male and female, work 3% fewer hours than their German counterparts. Part-time female employees in Greece only work 7% more hours than their German counterparts. (Data here, or see table below, which you can also download from here)
So that's the whole of the mainstream, more productive, labour market in a nutshell. The difference in hours worked on a weekly basis is nowhere near the staggering 48% implied by the OECD figures that Alex cites.
So what accounts for the difference in hours worked? For the most part it's the structure of the labour market. Fewer women and part-time workers in Greece, and a substantial fringe of workers that don't fit the traditional model of work.
The latter are very important. They are mostly part-time male employees, part-time non-employees of both sexes, and people with second jobs (data on those here). That's precisely the segment of the labour market that is shut out of fully-regulated work, not least by employment regulations. It's the kind of work, may I add, that our unions are, to this date, trying to regulate out of existence. Here, the difference is absolutely huge. A Greek man with a second employee job, for instance, works about twice as long as their German counterpart. Given the lower productivity in Greek workplaces, keeping working hours in mainstream jobs artificially low through employment regulation is not doing anyone any favours. A simple reading of Alex's data would suggest nothing's wrong. But there is plenty wrong.
There is much in this data that the European Left could use, if only they bothered to. The reading above is just one libertarian point of view. Getting to the bottom of why the Greek labour market includes a fringe working its metaphorical hairy knuckles to the bone is essential to understanding our plight. Unfortunately it doesn't play to the preferred narrative of evil neoliberals trying to talk the proud, untamed Greek worker down. Well the data aren't playing ball. What's it going to be, friends? Change the narrative, or manipulate the data? Do you want to win a flamewar, or do you want to understand the shit we're in? I'm not keeping my fingers crossed.
Bonus for returning readers: funny things happen when you compare like-for-like. Here's the evolution of average weekly working time for male full-time employees in Germany and Greece from 1983 to 2011:
And here is the same graph for full-time female employees:
UPDATE:
I've now finally been able to get the full European Values Survey dataset for all of Europe, which allows me to look exactly at what Greeks (and others) wanted from a job as of 2008. The full table (weighted data of course) can be found here. It shows that a greater share of Greeks than Germans want family-friendly jobs with low pressure and good hours, but a smaller share of Greeks than Germans needs jobs security. These comparisons were consistent even after filtering for only young people (34 or younger) or filtering for only unemployed people.
Is this conclusive? Well, much depends on what Greeks and Germans respectively call 'low pressure' and 'good hours'. I've partly divided the sample by old and young, employed and unemployed precisely to see if that makes a difference to perceptions. On balance I'd say the figures are a good enough indicator, but feel free to question below. Then there's the question of whether it's the Greeks or the Germans that are an outlier.
Also note that running figures by 'respondents' as opposed to 'responses' makes a big difference. In some countries, people focus on a few 'important' attributes of a job, while in others they cast a wide net. I will address this in an update soon.
If we want a single measure to compare how much people from different countries work, one idea is to combine working hours of different worker categories into a weighted average normalized by *population*.
ReplyDeleteThat is, if p1% of population works h1 hours per week, p2% works h2 hours per week, and so on, calculate the weighted average
h = p1% h1 + p2% h2 +...
As an example, if 25% of population worked 40 hours/week, another 25% worked 20 hours/week, and the rest 50% worked zero (0) hours/week, we would have an average of
h = 0.25*40 + 0.25*20 + 0.50*0 = 15 hours/week.
This measure has the advantage of taking into account unemployed, pensioners, and inactive individuals who work zero (0) hours per week. For example, it would account for the fact that "only 56% of Greek working age women were economically active in 2008, against 71% of German working age women".
My understanding is that the post you cite uses a weighted average normalized by *working* population. For the simplified example above, it would calculate
h' = 0.50*40 + 0.50*20 = 30 hours/week.
Ok, the way the article you criticize tries to show that Greeks are not lazy and that they are more hard-working than Germans is lame. It's a journalistic article, whatever data he found he put them there. But the main point, the fact that Greeks are not lazy and that they are more hardworking than Germans, is still valid. I would have written it like this:
ReplyDeleteFirst, we judge relative "laziness" and "hard work" by the amount of hours worked. I think that is the best we can do, considering that productivity measures capture so many things that in the end you can actually say nothing about the relative work effort across countries (just consider the different management, technology, industrial structure, capitalists' and unions' attitudes across the two countries). So, let's go on with the hours data.
Now, the OECD data are not appropriate, you are correct (alas, OECD has a disclaimer that says you cannot make comparisons across countries, you can only check the evolution in time!). Hence, we go on with the EULFS data. The points you make are valid, we should look at the structure of employment as well. But that does not change the fact that, on average, Greek workers spend more time in work than German ones.
Take only employees. This is the appropriate thing to do, since a self-employed person is a "capitalist-worker" mixture, so he/she cannot be included in the same sample (no academic study models and estimates their behaviour in the same way). Now, according to your data (and assuming that the part-time, full-time percentages hold also for employees - I know it's not correct, but that does not alter the argument), the "average employee" in Greece is a 92% full-time and 8% part-time worker and in Germany a 78% full-time and 22% part-time one. And also, the average worker is a "male-female" person. That's always the meaning of the "average", a non-existent person. The point about part-timers and full-timers and men and women is of course valid, but that does not change the fact that the "average" employee in Greece usually worked 5.3 (or 15%) more hours per week than the German one (39.8 versus 34.5 hours) in 2008, the highest amount in EU-15. 15% is a substantial difference of course (everything I report in this paragraph comes from your Eurostat bookmarked table, but augmented with more data and countries).
(continued...)
(...the rest)
ReplyDeleteBut we can also expand that and see the more complete picture (as the author of the article tried to do). Some crude calculations from the Eurostat database gave me a comparable percentage of workers holding a second job in Greece and in Germany. But when people hold second jobs, they work a lot more hours in Greece than in Germany (as seen from your data). This raises the "average" amount of hours Greeks work and contributes positively to the above difference. Moreover, Greeks are much less likely than Germans to be absent from work. There is plenty of evidence on that (check http://www.imf.org/External/Pubs/FT/staffp/2007/03/pdf/lusinyan.pdf and http://mpra.ub.uni-muenchen.de/22627/1/MPRA_paper_22627.pdf). This again contributes positively to the difference. Also, public (or "bank") holidays are almost the same (at least as I can infer from this http://www.123newyear.com/bank-holidays/2011/europe.html). But legal holidays and vacations are much more generous in Germany (here, Fig.1 http://www.cepr.net/documents/publications/2007-05-no-vacation-nation.pdf). And parental leave as well: http://en.wikipedia.org/wiki/Parental_leave, http://www.eurofound.europa.eu/pubdocs/2006/87/en/1/ef0687en.pdf). All these things add to the annual perspective of working hours and contribute positively to the Greek-German workin time differential.
So, that's how I would have made the case that "Greeks are not lazy and they are more hard-working than Germans".
I hope you 'll enjoy my input!
M. Veliziotis
PS. I could have added various "quality" indicators as well from the European Working Conditions Survey, related to working time, but that would have made the comment really and unecessarily long.
4X as many ppl don't work at all in Greece(as in Germany).
ReplyDelete@anon yes but they are not included in working time stats. Also, remember to use 2008 figures
ReplyDelete@anon yes but they are not included in working time stats. Also, remember to use 2008 figures
ReplyDelete@anon @Manos but overall employment level may be important as well. If employment levels are lower in Greece (which I think they are), then those who do work, ceteris paribus, need to work more in order to produce the same output.
ReplyDeleteLet's consider this: assuming no diminishing returns on hours worked, if half of actual number of Greeks in 2008 worked for ca 4000 hrs/year, they would produce the same output!
Is it possible that the average Greek worker (this amalgam of full/part time, male/female, first/second job people) works more hours because on average he/she has to support more non-workers?
@kalamake it's not a bad interpretation, but it's trying to make too many arguments at once. Full time workers and female part-time workers, as I've shown, work almost exactly the same hours in Greece and Germany. Presumably they have as many dependents, on average, as the others. So either the fringe workers are forced by employment law and taxation to subsidise the full-timers, or something else is at play. I don't want to deny your point, but the kind of argument I'm trying to make (i.e. that it's the structure of our labour market that's wrong, not laziness or the exploitative nature of the employee relationship), is easier made by comparing like for like.
ReplyDeleteI think this discussion is somewhat misguided, first off you don't need to analyze the OECD data by some kind of complex processing to see that they have little bearing on the real situation. This is because the laziness most people allude to doesn't have to do with hours worked per month but with the fact that like 15%+ of the jobs in the greek public sector essentially contribute nothing to the GDP or society. You could be some middle aged guy at the IRS that just signs 10 papers per day, or a middle school teacher that is allocated to teach 15 hours per week, or some orthodox priest that goes to some remote church once every 20 days, or a municipal cop that gets 1000 euros per month to spend 4 hours per week watching over the kids renting bikes at the park, or even worse some politically involved b**tch with a liberal arts degree that gets a management position in one of those MKO's with 3000 euros per month and a chance to steal more, and of course the mandatory greek syndicalist that might even go and sabotage his freaking equipment in order to ensure that he can spent the entire day dealing with politics instead of doing something productive.
ReplyDeleteAll of the above claim they work 8 hours a day , cause hey in greece you can claim anything you want at all. Those employee evaluation tests that supposedly ensure "meritocracy"? Every single employee gets 10 out of 10. And ironically they might even be tied to salary bonuses so instead of being a way to give incentive to productive people they end up as yet another way for the politically involved to get free money.
But the real criticism to those debunking the myth of the "lazy greek" has to do with the fact that in the first place, greeks aren't really accused of sitting all day and drinking coffee or something, but of not having the ability to organize their asses in to doing something that isn't rudimentary and that requires planning ahead instead of the division of people in political nepotism and clientelism groups that only waste their time examining ways with which to secure money for themselves first.
I mean there are hard working greeks that work for 15 hours a day, but they are almost always self employed or working at some other small scale place (carpenters, restaurant owners, some in the clothing and textile industry and many more), but the thing is outside the small businesses you are gong to get lynched but the mixture of clientelism and corruption of the "antineoliberal" mentality of the greek system and the people working for it.
You are absolutely right. One other piece of data that I'm sure is overlooked is the fact that almost every public sector workers would actually would clock so much overtime that the government once calculated that it would actually save millions if it was to send its workers home and just pay them their salaries without overtime. Think about it there are over 750.000 public workers.
DeleteI know one thing. I am self employed I don't even know how many hours I work a week but the funny thing is no one has asked me. So I know that the stats don't take people like me into account. So they are bogus. Furthermore, I really don't think that I'm anymore special than any self employed worker in the entire world. That's why we do what we do - we are workaholics - but that like I said is true whether your Greek, German, American or Italian. The self employed are not the cause of the problems the world is facing now - we are the solution. The government on the other hand...