7 ms·

It's fascinating to me that even august publications like the Wall Street Journal continue to use 'unemployment rate' as any kind of reliable metric, and says s

It's fascinating to me that even august publications like the Wall Street Journal continue to use 'unemployment rate' as any kind of reliable metric, and says stuff like 'the unemployment rate is at a 50 year low now'. Unemployment just measures those who are actively looking for full-time employment but are unable to find it (and yes there is a separate underemployment metric, which seems a little more accurate). The reason we have 'slack' in the labor market is because the adult prime-age participation rate in the economy was quite low for a long time after 2008, and even now is only getting back to early 2000s numbers. This just seems like a much better metric to understand the health of the labor market & perceived slack. Disability has exploded over the last 20-30 years, and someone collecting a disability check- especially if they are simply milking the system- doesn't show up as 'unemployed' because they're not actively looking for work. It's actually pretty concerning for the US how low the adult participation rate has fallen. There's pretty good evidence that this decade of low interest rates has started to slowly bring in people off the sidelines who weren't looking for work, had given up, were in the informal sector and weren't counted, etc. That's a massively good thing for society. I don't think you can understand labor market slack or the point at which wages will start to rise until you ditch the somewhat gimmicky 'unemployment rate' and start looking at adult participation in the economy period

bhupy 7y ago
The U6 rate is also at historic lows, and that's the broadest definition of "unemployment", which includes discouraged workers. https://fred.stlouisfed.org/series/u6rate https://fred.stlouisfed.org/series/u6rate
- coldtea 7y ago
  What they never measure is the big rise in BS near-substinence hand-to-mouth jobs over higher quality gigs - which is the case in many (most?) countries...
  - bhupy 7y ago
    You're right, that information is captured by the median weekly inflation-adjusted earnings for wage and salary workers, which is actually at historic highs. https://fred.stlouisfed.org/series/LEU0252881600A https://fred.stlouisfed.org/series/LEU0252881600A Combining the U6 unemployment rate with the real median earnings paints a more complete picture.
    - lotsofpulp 7y ago
      I would have to combine it with changes in CoL in areas of economic opportunity versus those without. Parts of the country offer far more security in future income than others.
      - rayiner 7y ago
        That would penalize the national statistics for the dysfunctional laws and policies of a handful of localities.
        lotsofpulp 7y ago
        How else can one account for increased volatility in labor markets due to automation and outsourcing?
    - lazulicurio 7y ago
      Genuine question: looking at that graph, it's saying that median weekly earnings reached a peak in 2009 then dipped 2010 through 2013. I find it rather counter-intuitive that a peak in median earnings would occur during the recession. What would be the explanation behind that peak?
      - finebalance 7y ago
        Possibly: low income earners lost jobs in substantial enough numbers to affect the median (that to some extent is robust to outliers).
rayiner 7y ago
Labor market participation rate is not a great measure. Look at the breakdown of changes in labor market participation by age: https://www.bls.gov/emp/tables/civilian-labor-force-participation-rate.htm https://www.bls.gov/emp/tables/civilian-labor-force-particip... From 1998 to 2018 participation dropped just 2 points for those age 25-54. That is projected to stay stable through 2028. It went up for those 55+. The only place it dropped significantly was 16-24. I.e. our society is richer and we can afford to have more young adults spend their early 20s binge drinking instead of working. As to disability—it’s hard to tell whether increasing numbers of people on disability is good or bad. If the government has simply gotten more generous and allowing more people with real problems to draw disability, that’s not a bad thing.
- refurb 7y ago
  This is a great example of why you should dig into the numbers behind a statistic. Everyone claims labor participation is still low and throws around their pet theories as to why, yet if you simple dig into the numbers like rayiner did, you’d realize most of those pet theories are wrong.
- hash872 7y ago
  FRED seems to have some different results https://fred.stlouisfed.org/series/CIVPART https://fred.stlouisfed.org/series/CIVPART. These are huge numbers in a country the size of the US. Anyways, 'projected' is meaningless. It's pretty widely acknowledged that disability is abused, it seems to be a mix of people with real problems and those without. The very high disability rates in a few counties I think shows that it's not always legitimate usage
  - rayiner 7y ago
    The FRED data is identical and probably the same source. It shows labor force participation rate dropping from 67% in 1998 to 63%, same as the BLS data I posted. But the chart you linked to doesn’t break things down by age group. When you do, you can see that the drop is almost entirely due to the 16-24 age group. Participation for people 25+ is stable since 1998.
paulddraper 7y ago
> use 'unemployment rate' as any kind of reliable metric All metrics are flawed, and each has their use. The unemployment rate helps you understand how likely the average worker is to find a job if they look for one. Obviously many potential workers will choose not to: health, wealth, sloth, depression, age, wages. But one aspect of a functioning economy is being able to match willing workers to jobs. For example: Though I don't put much stock in the "replaced by robots" forecasts, if it were to happen, the unemployment rate would rise.
throwaway743 7y ago
What you'll never hear is how the bottoming of unemployment has preceded the last 9 or so market crashes. So it's quite the opposite in terms of being an indication of the market's good health.
- ramblerman 7y ago
  That's a bit like saying, the patient seemed to be in good health right before he got sick the last 9 times.

bhupy 7y ago

The U6 rate is also at historic lows, and that's the broadest definition of "unemployment", which includes discouraged workers. https://fred.stlouisfed.org/series/u6rate https://fred.stlouisfed.org/series/u6rate

coldtea 7y ago
What they never measure is the big rise in BS near-substinence hand-to-mouth jobs over higher quality gigs - which is the case in many (most?) countries...
- bhupy 7y ago
  You're right, that information is captured by the median weekly inflation-adjusted earnings for wage and salary workers, which is actually at historic highs. https://fred.stlouisfed.org/series/LEU0252881600A https://fred.stlouisfed.org/series/LEU0252881600A Combining the U6 unemployment rate with the real median earnings paints a more complete picture.
  - lotsofpulp 7y ago
    I would have to combine it with changes in CoL in areas of economic opportunity versus those without. Parts of the country offer far more security in future income than others.
    - rayiner 7y ago
      That would penalize the national statistics for the dysfunctional laws and policies of a handful of localities.
      - lotsofpulp 7y ago
        How else can one account for increased volatility in labor markets due to automation and outsourcing?
  - lazulicurio 7y ago
    Genuine question: looking at that graph, it's saying that median weekly earnings reached a peak in 2009 then dipped 2010 through 2013. I find it rather counter-intuitive that a peak in median earnings would occur during the recession. What would be the explanation behind that peak?
    - finebalance 7y ago
      Possibly: low income earners lost jobs in substantial enough numbers to affect the median (that to some extent is robust to outliers).

rayiner 7y ago

Labor market participation rate is not a great measure. Look at the breakdown of changes in labor market participation by age: https://www.bls.gov/emp/tables/civilian-labor-force-participation-rate.htm https://www.bls.gov/emp/tables/civilian-labor-force-particip... From 1998 to 2018 participation dropped just 2 points for those age 25-54. That is projected to stay stable through 2028. It went up for those 55+. The only place it dropped significantly was 16-24. I.e. our society is richer and we can afford to have more young adults spend their early 20s binge drinking instead of working. As to disability—it’s hard to tell whether increasing numbers of people on disability is good or bad. If the government has simply gotten more generous and allowing more people with real problems to draw disability, that’s not a bad thing.

refurb 7y ago
This is a great example of why you should dig into the numbers behind a statistic. Everyone claims labor participation is still low and throws around their pet theories as to why, yet if you simple dig into the numbers like rayiner did, you’d realize most of those pet theories are wrong.
hash872 7y ago
FRED seems to have some different results https://fred.stlouisfed.org/series/CIVPART https://fred.stlouisfed.org/series/CIVPART. These are huge numbers in a country the size of the US. Anyways, 'projected' is meaningless. It's pretty widely acknowledged that disability is abused, it seems to be a mix of people with real problems and those without. The very high disability rates in a few counties I think shows that it's not always legitimate usage
- rayiner 7y ago
  The FRED data is identical and probably the same source. It shows labor force participation rate dropping from 67% in 1998 to 63%, same as the BLS data I posted. But the chart you linked to doesn’t break things down by age group. When you do, you can see that the drop is almost entirely due to the 16-24 age group. Participation for people 25+ is stable since 1998.

paulddraper 7y ago

> use 'unemployment rate' as any kind of reliable metric All metrics are flawed, and each has their use. The unemployment rate helps you understand how likely the average worker is to find a job if they look for one. Obviously many potential workers will choose not to: health, wealth, sloth, depression, age, wages. But one aspect of a functioning economy is being able to match willing workers to jobs. For example: Though I don't put much stock in the "replaced by robots" forecasts, if it were to happen, the unemployment rate would rise.

throwaway743 7y ago

What you'll never hear is how the bottoming of unemployment has preceded the last 9 or so market crashes. So it's quite the opposite in terms of being an indication of the market's good health.

ramblerman 7y ago
That's a bit like saying, the patient seemed to be in good health right before he got sick the last 9 times.