This is the case anytime there’s a dip in employment - people have more incentive to apply for disability benefits, so the rolls go up. It’s certainly not a direct relationship between people who become disabled and then apply - there are other inputs, and drawing conclusions about the unemployment rate from disability benefits application numbers is a little misleading, since each affects the other.

By the way there’s typically a nine month wait between an application for disability benefits and a decision, so I wonder whether the government is counting applicaants or recipients… it would be strange not to count applicants, since the application process is terminated as soon as you go back to work.

In Europe, there is a well-trodden path for window-dressing politically embarrassing unemployment rates by easing the eligibility criteria for disability benefits. I recall from years back, some illuminating Financial Times features on just this, which are inaccessible now except, perhaps, for those who have a subscription.

The short answer is that we need to look at national participation rates of the working age population in employment, not just the official standardised unemployment rates, which tend to make the media headlines. Try Chart 5 in: http://europa.eu.int/comm/employment_social/employment_strategy/pdf/etf_annex2_en.pdf

The EU Labour Force Survey for 2002, published in 2003, is at:
http://www.eu-datashop.de/download/EN/inhaltsv/thema3/arbeitsk.pdf

Other useful links at:
http://www.unece.org/ead/pub/022/022_4.pdf
http://www.ecb.int/pub/pdf/labourmarket2002.pdf

The Scandinavian countries, the Netherlands and the UK come out well, with relatively high participation rates.

“This is the case anytime there’s a dip in employment - people have more incentive to apply for disability benefits, so the rolls go up.”

This is obviously true, the curiousity here is that this tendency now seems to be continuing and possibly accelerating accelerating now that the US is supposedly in recovery mode. Viz the 400,000 people who have apparently recently dropped out of the US labour market according to last Friday’s numbers.

Well, you would certainly expect a lag, as I explained above. But the idea that the US is in recovery mode in terms of job is a little sketchy… just last week the job figures came out and they were much worse than expected. So far yes we are experiencing a recovery - but it’s a jobless recovery a la the early 90s.

Goolsbee is overstating the impact. Based on his 1,000,000 number (which is also probably too large based on a reading of Autor and Duggan) to the BLS data on the unemployed and labor force and unemployment would have peaked at 6.9%, not 8%.

Actually reading the research and checking the data is something I’d expect from a Prof. at the Univ. of Chicago.

Bob,
Interesting, that chart 5.

A similar chart that Brad DeLong produced for the U.S.:
http://www.j-bradford-delong.net/movable_type/2003_archives/003026.html
(scroll down a bit)
…shows U.S. Employment as a ratio of the working age population shrinking perhaps 2% between 97 and 02, from almost 64% to a bit above 61% (a worse employment rate than the EU15 as a whole).

The trend for the U.S. employment/population ratio in the last three years does not show anything that could be considered a recovery.

Patrick,

I tend to get nervous about making cross-country comparisons with data, and especially with graphs, unless I feel reasonably confident the data are all on standardised basis. If national populations of working age are a clear enough notion, national employment data may be far from it. Differential treatments of part-timers, self-employment and non-civilian employment leave much scope for wobble across countries so I tend to look for standardised OECD data sources when comparing the EU with North America.

Unhappily, I’ve not found data for general employment participation rates across the OECD for a more recent year than 2001: http://www.oecd.org/dataoecd/8/4/1874420.pdf

However, I did come across this very recent working paper on female participation rates, which could be of interest to a passing reader:
http://www.olis.oecd.org/olis/2003doc.nsf/43bb6130e5e86e5fc12569fa005d004c/b7c9b45202b081b1c1256e0000317f04/$FILE/JT00155820.PDF

By way of a benchmark, I came across this from OECD Observer:

“In 2002, the proportion of the population of working age who had a job in Canada and Denmark was, respectively, 72% and 77%. These figures are higher than those in the UK and the US, which can be regarded as good performers…However, it is interesting to note that the employment rate among unskilled workers in Canada and Denmark, at practically 55% and 63% respectively, is also relatively high.” Sadly, there were no data for other countries.

I take Edward’s point about the acceleration in the decline of the US employment ratio indicating that there is no sign of an upturn in the job market there. In the Eurozone, the big immediate business worry is the appreciating Euro against the US Dollar and that is when employment rates in the major Eurozone economies were rather on the low side compared with the US.

Ever since reading that the growth of average real earnings in the Netherlands during the 1980s was kept at about 1% a year, I think we have to look at the smaller economies differently from the large ones. It is sometimes possible to operate a de facto incomes policy by consensus in a small economy when that is unthinkable in a large one. Yet another reason for being cautious about some EU sourced analysis is that I occasionally get a feeling the analysis is driven to score points against the US or the “Anglo-Saxon” countries.

Bob,
Those two graphs aren’t useful for anything but the most cursory of analyses:
(1) that u.s. employment/working age pop. is within the range of European numbers. If perhaps on the low end. Definitely not head and shoulders above Europe.
(2) That the U.S. recovery of 2001 is a mirage.

That OECD paper you found with the “2001” data, seems to have a different definition of workforce participation whose numbers do not make it comparable to either graph.

The growth in American disability pay out rates has less to do with “book cooking” than it has to do with American welfare reform.

The end of permanent welfare benefits resulted in large numbers of people moving over to Supplimental Social Security disability pay outs.

In short, we are seeing an intergovernment cost shifting game between local and state government on one side and the Federal government on the other.