June 08 Labour market: Part Two (Revenge of the hours worked)
This follows from Part One.
Well, the unemployment rate rose to 3.9% – this was above market expectations, so the immediate feeling would be that the labour market is in trouble.
However, then you see the increase in employment and HOURS WORKED (the indicator I wanted to keep an eye on) and you realise “this employment data isn’t hot – but it is a definite improvement on what last quarter implied”.
So what am I talking about, and why am I talking in the third second person? (ht CPW) Well lets start with the first question, and lets look at the hours worked numbers.
Note: Other blogs on the numbers (Rates Blog) (The Hive) (The Standard) (Kiwiblog)
The hours worked story
During the March quarter hours worked fell 1.7% in the household labour force survey, but they didn’t in the quarterly employment survey. People tend to trust the HLFS more and so they got scared. Of course, it turns out that the HLFS has decided to be jumpy as hell, with hours worked rising 2.4% over the June quarter – more than reversing the March fall.
As a result, hours worked has not tumbled as we would expect at the beginning of a recession – now this in no way implies that we aren’t having a recession. What it does imply though, is that the chronic shortage of labour that we have had in the past has helped prevent much loosening in the labour market in the face of a recession – a fact that must stress the hell out of the RBNZ, which is why the dollar lept when the numbers came out.
My interpretation is that today’s number, by itself, overstates the strength of the labour market. However, the March number, by itself, understated it. I think that the labour market has stalled. It makes sense that firms will want to hold labour – given the difficulties they have had getting them in the first place.
Furthermore, the fact that hours worked haven’t fallen substantially on a year earlier is very interesting – and seems to imply that demand for labour by firms still exists, a piece of information that is just not consistent with firms that anticipate a prolonged slowdown!
One extra tidbit – employment and hours worked rose during a period of massive structural layoffs. Silver Fern was closing down plants as it restructured and Fisher and Paykel decided to produce overseas. With all these anecdotes of a collapsing labour market we didn’t get one – what gives? If anything this suggests to me that the labour market may be even STRONGER than current numbers suggest – interesting.
Third Second person?
Now why was I talking in the third second person? Well there are two reasons:
Firstly, I suspect that these numbers will force the RBNZ to ask itself some tough questions – as they used the weak March numbers to justify cuts, and if anything today’s result implies that the March numbers were spurious. By going into the third person I can try to imagine what they must be thinking.
Secondly, I’ve had too much coffee, and didn’t get much sleep last night after seeing the Phoneix lose the pre-season cup final – damn the boys deserved a win!
Conclusion
When this result is taken in conjunction with the QES numbers, the RBNZ must be starting to feel nervous about inflation!
“you see the increase in employment and HOURS WORKED (the indicator I wanted to keep an eye on) and you realise “this employment data isn’t hot – but it is a definite improvement on what last quarter implied”.”
That’s the second person actually. 😛
“That’s the second person actually. :P”
Flip – I thought it might be, but I thought stating third person was the safe option 😉
Damn you and your ability with english 😛
I’d expected continued drops in female labour force participation with WFF; I was wrong. If you look at Table 10 on the spreadsheet, you get work status (both working, mixed, neither working) for couples by whether they have children and how many children they have. Now, a proper study would segregate things by income and WFF eligibility, but I’d at least expect in the aggregate numbers that we’d see overall rises in the proportion of mixed-income couples with children relative to two-income couples with children. If you take those ratios, there’s not really much action over time. There’s an increase in the fraction of families with 3 or more kids who fall into mixed work rather than both working, but I’m not sure we can distinguish it from noise. I’d still love to see a proper analysis that would have Table 10 data tabulated by income cohort but it looks less likely that WFF has had big effects of female labour force participation.
“it looks less likely that WFF has had big effects of female labour force participation”
To be fair, we have had a period of extremely strong income growth – so the return from working for secondary earners has been rising in unison with the negative effect of working for families.
Ultimately, labour supply is fairly elastic for secondary earners, so it is easy to imagine a situation where the response to WFF is clocked by the increase in labour demand over the past few years. Once we’ve seen a full economic cycle the policies impact should be more transparent.
Oh, I’m not saying it’s a settled issue at all. And especially not without the disaggregated data. But, I’ve revised downwards somewhat my expectation that it’s mattered a lot.
Regardless of strong income growth, the return to working for secondary workers remains more attenuated the more children you have (with effective marginal tax rates hitting 80% or so for some cohorts). If WFF mattered a lot and we had a strong income growth period, we’d then expect a decreasing fraction of couples without children to have only one earner and an increasing fraction of couples with many children to have only one earner. If you take the ratios, there’s no movement really at all in the ratio of 1-worker:2-worker childless couples and, among those with kids, there’s not really any movement in that ratio except among those with 3+ kids.
But wouldn’t the fact that someone is in a position where they are eligible for WFF increase their reservation wage, which in turn would increase their bargaining power, which in turn could lead to them receiving a higher wage than people who are not eligible for WFF – given the tightness of the labour market.
In such a case, we wouldn’t expect to see much change in the ratios because the burden of the tax would fall on the employer. However, it should show up in income growth for families with children vs families without.
Well, it’s true that a negative labour force supply shock will mean that the reservation wage of the marginal worker will be higher, but I can’t see how that pushes up wages for the cohort with the particularly higher reservation wage rather than pushing up wages overall and increasing the proportion of those with the higher reservation wage that are out of the workforce. Maybe you could specify a model where some jobs really can only be filled by workers who have experience with parenting, but that just seems hokey.
We would expect the most action among those with 3+ kids, and that’s where there are hints as to action in the data, albeit noisy.
Would be interested in seeing a longer time period than the one in the hlfsjun08 dataset that Stats put out…that one truncates at June 06. It doesn’t look to me like there’s much action in the current time period, but maybe it’s there if we compare it to a longer time period.
“but I can’t see how that pushes up wages for the cohort with the particularly higher reservation wage rather than pushing up wages overall”
I guess I was implicitly assuming that the ability to receive WFF was observable by employers – that might not be a realistic assumption 😛
Even if it’s observable (and I’d think it is), it can’t lead to more than transitional frictional gains for eligible folks: long run, they’re replaced by folks with the lower reservation wages.
Also just to be clear, I was assuming that the pool of non-child workers had been used up – which is obviously not the case in the graph you just email 😛
I don’t have the data on me at the moment, I’ll have a check on infos later on – the tables don’t have infos identifiers though
Ahh you replied before I replied 😛
Yes, you’re right – ultimately it will drive up everyone wages as the WFF worker is the one at the margin
Conclusion:
Employers screw ever more hours out of workers lives because labour is so cheap.
http://rogernome.blogspot.com/2007/10/another-reason-new-zealand-needs-its.html
“Employers screw ever more hours out of workers lives because labour is so cheap.”
These increase are in paid hours – people are getting paid to work more. Employers aren’t chaining them in the workhouse and making them do more, people are trading these hours for a greater claim on resources.
Would you prefer it if the work and the additional pay that comes with it wasn’t available for people?
“Would be interested in seeing a longer time period than the one in the hlfsjun08 dataset that Stats put out…that one truncates at June 06”
Hi Eric,
The series isn’t on Infos – if you email Stats NZ I’m sure they’ll be able to send you a backdated version of table 10.
It’s more something that I’m keeping an ongoing eye on — if it starts to look like it could be interesting enough, I’ll see about getting a CURF that would let me look for effects on different income/family size cohorts.
How come we spend so much on (eg) Telecom mobile, Vodaphone (her), broadband (Telstraclear) line phone…. ???
“How come we spend so much on (eg) Telecom mobile, Vodaphone (her), broadband (Telstraclear) line phone…. ???”
Well telecommunications networks probably have increasing returns to scale going on – which implies that areas with larger populations will get lower prices.
Actually, a more convincing story is that the more dense the population it is, the lower the average and marginal costs of servicing the area – as a result of our lower population density you would expect prices to be higher in NZ.
That would be my guess anyway 😛
Matt – some choices are “clayton’s choices” – i.e. you don’t really have much of a choice when it’s “work extra hours or lose your job”. This is one of the many reasons why neoclassical labour market theory isn’t applicable to reality.
“Would you prefer it if the work and the additional pay that comes with it wasn’t available for people?”
No. I would prefer overtime to be available at time an a half, like it is in most OECD countries.
“i.e. you don’t really have much of a choice when it’s “work extra hours or lose your job””
How many jobs are actually like that? Even if they are – there is a choice to leave the job. When we have a tight labour market, as we currently do, this choice offers a higher reservation payoff than just the unemployment benefit. Firm’s still have to AT LEAST meet this reservation level – so what is the problem?
“No. I would prefer overtime to be available at time an a half, like it is in most OECD countries.”
Now I’m not arguing with that – if we believe there is a negative externality associated with “too much work” overtime allows us to tax this externality (note I am not saying I support it either).
However, the data isn’t even saying that – overtime hours fell! People reporting that they were “under-employed” (eg would work more hours) is still at the same level it has been for a while. This data does not say what you did which was:
“Employers screw ever more hours out of workers lives because labour is so cheap.”
That is what I’m disputing.
“How many jobs are actually like that? Even if they are – there is a choice to leave the job.”
I was actually being a bit lazy with that one. It’s as much about getting off-side with your employer as it is with possibly losing your job.
When I worked for a furniture moving firm, i was working 45-50 hours between monday and friday, and then the boss would often demand that I work saturday, and sometimes sunday. This went for many of the other workers as well. If I refused to work the 50-60 hours per week that he wanted me to work he would harass me for the rest of the week – which was extremely uncomfortable, so I just gave in and worked the extra hours. I can’t imagine how this would work if I had children, etc…
I had friends in the workplace and was well established there so didn’t really want to re-establish somewhere else etc… You see, this is where the theory just doesn’t concord with reality – humans are social beings, and don’t always fit within the assumptions of algorithms.
“if we believe there is a negative externality associated with “too much work”
Clearly there is. There is a social cost, especially when you have a family that you want to be with in the weekends. But this is where right-wing economists usually fail – social costs are less important than “economic efficiency” or “growth”, because they don’t fit nicely into your little equations. They are hard to quantify and therefore frustrate you, so you discard them as unimportant.
“Employers screw ever more hours out of workers lives because labour is so cheap.”
If employers were taxed for the externalities involved of over-working their employees, employees would have to work less hours for the same amount of money. Are you really suggesting that most employees wouldn’t prefer that situation? If not, then clearly employers who demand employees work longer hours are certainly making employees work longer hours than they would prefer.
“humans are social beings, and don’t always fit within the assumptions of algorithms”
Roger, if you actually want a civil discussion about these issues you need to stop going on about economists in this way. Have a derived a single equation in this post. Have I ever used maths on this blog outside of accounting identities and they odd logic check? Constantly pestering us about using maths (as a logical check of ideas – which is a good way of using it) isn’t going to get us anywhere.
I do realise that there are asymmetries in the bargaining process. Government can help this to a degree – and I think it currently does. Economists do not ignore this sort of thing, no matter how much you tell yourself that they do.
“But this is where right-wing economists usually fail – social costs are less important than “economic efficiency” or “growth”, because they don’t fit nicely into your little equations”
Again that is rubbish. For one, they do fit nicely in our mathematical representations.
Secondly, you call me a right-wing economist, which implies you think I ignore social costs.
However, I CONSTANTLY propose that we should have things like a fat tax, or a petrol tax, in order to cover social externalities – hell I was even the one that suggested the externality idea for this post!
Economists do not think growth is the most important thing – they think that being in a situation that provides the most happiness for society is the most important thing. That is why I stated that I wasn’t concerned about a current account deficit in a later post that you have commented on – if all I cared about was growth I would hate the current account deficit.
“They are hard to quantify and therefore frustrate you, so you discard them as unimportant”
If you read much modern economic literature you would realise that this statement is wrong – I’m sorry but you are attacking economics instead of attacking the fundamental ideas, and you are mis-representing economics in the first place so that you have a straw man to attack.
“If employers were taxed for the externalities involved of over-working their employees, employees would have to work less hours for the same amount of money”
No, they would be paid less. The cost of an employee would increase, but the share going to the employee would fall.
Now, if you cut out all the lingo out “right-wing economists” and “mathematical models” I’m sure that you would see that we agree on large portions of what is going on here. We might just have different subjective beliefs surrounding the size of any “income externality” in society. That is fine – subjective beliefs are based on experience, we are allowed to differ and should both stand up for these beliefs. However, you are not doing this case any favours with your lines of attack.
“No, they would be paid less. The cost of an employee would increase, but the share going to the employee would fall.”
But Matt – from 1991 (when controls on no-standard work began to be dismantled) to 1996 (when they were all but gone), the median hourly wage rate fell, as did the median weekly income. In those five years GDP per capit grew by nearly 20 percent, and negative wage gains. Business profits grew exponentially though
Here’s a bit of reality, my theoretically focussed friend:
So what the hell are you basing your claim on?
When talking about taxing an “externality” Roger said – “employees would have to work less hours for the same amount of money”
Then when I disputed that Roger said “So what the hell are you basing your claim on?”
The identity that pay=hours*hourly rate – and you have said that they would get paid more when working less when the employer is taxed for making them work.
Now if hours are lower, and the hourly rate is the same (as taxing overtime, or taxing all hours would not increase their net hourly pay) it doesn’t imply that there pay will be greater.
Your evidence does not defend what you said at all.
“Here’s a bit of reality, my theoretically focussed friend:”
Do you realise that I spend all day, about 6 days a week, going through economic numbers, checking samples, checking survey questions etc. Most applied economists do. I still don’t understand why you don’t think we pay any attention to the data?
“When talking about taxing an “externality” Roger said”
Please don’t address me in the third person. We’re having a conversation here.
“Now if hours are lower, and the hourly rate is the same”
But if penal rates are applied, the hourly rate is higher. Sharpen up man.
“Your evidence does not defend what you said at all.”
Oh but it does. The reduction in penal rates was one of the key reasons that real wages fell under the ECA. Why is this so hard for you to understand?
“Do you realise that I spend all day, about 6 days a week, going through economic numbers”
Why do you persist with ideas that have so little basis in reality then?
“Please don’t address me in the third person. We’re having a conversation here.”
Sorry, I only did that because I was taking lines from old comments – and I didn’t want other people getting confused.
It is more than a conversation – it is also something for other people to read 🙂
“But if penal rates are applied, the hourly rate is higher. Sharpen up man.”
Penal rates are for specific days (such as weekends) – not for the whole pay period. As a result, having penal rates IN NO WAY indicates that people will get paid the same or more to work less hours – your argument is not logically consistent.
“Oh but it does. The reduction in penal rates was one of the key reasons that real wages fell under the ECA. Why is this so hard for you to understand?”
I didn’t say anything about the ECA or what happened to real wages – you are changing the point again to make it sound like you are right! You said the re-introduction of penal rates would lead to people working less hours and getting paid the same amount or more.
This depends on both:
A) The elasticity of demand for the penal hours,
B) The proportion of total hours made up by penal hours.
In most cases this won’t be sufficient to ensure that people are paid the same to work less hours.
Now I suspect you might say “arrgg you used theory, which means what you say is irrelevant”. However, that shows an inability to think on your part – it does not indicate that you have a superior knowledge of “reality” to me.
Again, theory is a way of organising ideas – as a student I thought you might have to do this occasionally.
“Why do you persist with ideas that have so little basis in reality then?”
Just because what I say doesn’t conform to your view of reality does not mean it isn’t correct, or based in information. Try to stop sounding so arrogant.