« Where has all the research gone? | Main | An erratum on US employment flows during the recession »


Feed You can follow this conversation by subscribing to the comment feed for this post.


Are the LFS data amenable to matching? I've used the CPS files to get some panel dimension on the respondents before. I feel like you must be matching, unless the LFS asks some fairly specific questions which you wouldn't find in the American CPS.

No, you can't track the same respondent from one month to the next using the public use files. Apparently you can with the master files. I'm getting my courage up to try and do that.

The story of the US recession was not one of mass layoffs; the flows of layoffs are fairly stable throughout.

Really? Are you sure you aren't the victim of an optical illusion Nick? In the months from January 2007 through mid-2008, the monthly layoffs and discharges number appears to be around 2 million. But in the months right around January 2009, that number seems to go up to around 2.75 million to 3 million. A 40% to 50% uptick in dismissals seems to me to qualify as a mass layoff.

Also, those are the raw numbers. Since the workforce was shrinking, then even if the monthly layoff stayed constant, layoffs as a proportion of the labor force were rising dramatically.

Dan: that post was by Stephen, not me. (Stephen's the one who is good with numbers and charts and graphs and things).

Stephen, there is a public use file for the SLID (survey of labour and income dynamics), which is the longitudinal LFS - that would allow you to track respondents. I haven't used it, however. (Now back to marking).

Frances: I'll take a look at that. Offhand, do you remember if SLID comes with weights that would give estimates of how many people experience a given transition in a given month?

Dan: You're quite right - I was just eyeballing that JOLTS graph. But I just checked the raw numbers and yes, there was a jump in layoffs during roughly the same time as here. I'm going to do a follow-up post.

Is there any breakdown by occupation or sector?

Are any of these numbers linked to exports to America, e.g. timber products or transportation for export, etc.?

It's possible, yes. Just a matter of taking the time to extract it.

I know that studying labour market transitions is old hat for for labour economists, but there seems to be business cycle aspects that are not well-understood.

Stephen: "Offhand, do you remember if SLID comes with weights that would give estimates of how many people experience a given transition in a given month?"

I'd be very surprised it if didn't have weights, pretty much every PUMF that I've ever looked at does. Unfortunately, I don't find them particularly easy to use, and whether or not they would be reliable enough to give good estimates of the number of transitions is another matter entirely. Kevin Milligan would be the person to ask on this. Typically I've seen SLID used for estimating the probability of an individual person experiencing a particular type of transition, e.g. hazard models for the probability of exiting unemployment.

Yes, I've seen lots of those sorts of studies. I just checked Equinox, and the latest SLID PUMF is only from 2006. Is that normal? Does it take that long for the files to be prepared?

Stephen, odesi, which is the Ontario University data portal, has up to 2009, and I'd be astonished if Laval didn't have equally good access through the data liberation initiative.

I don't think you'd be able to get SLID data that's recent enough to look at, e.g., this year's macro trends. There are huge confidentiality issues with longitudinal data so it takes a while to be cleaned up. Also the policies around release of longitudinal data seem to be a little inconsistent, e.g. the first three waves of the National Longitudinal Survey of Children and Youth are available, but later waves aren't (it's amazing how much I've learned in this year's micro honours research seminar, 2 years ago I wouldn't have been able to tell you any of this!)

Its so unsettling when the main wage earners of households lose their jobs, possibly meaning that the younger or older members of the household may need to look for work just to avoid getting into serious debt. Not a good situation.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad