5 ms·

Correlation is not causation (fighting urge to post XKCD comic). "Unemployment fell, most recently to less than 4 percent, more than a full percentage point low

by sytse 10y ago

Correlation is not causation (fighting urge to post XKCD comic). "Unemployment fell, most recently to less than 4 percent, more than a full percentage point lower than the national rate." The fall can be explained by an improving economy. Unemployment was 4.2% on Dec. 31, 2013 https://ycharts.com/indicators/seattle_wa_unemployment_rate https://ycharts.com/indicators/seattle_wa_unemployment_rate National rate was 6.7% at that time. So the difference between Seattle and national has been reduced, they are relatively worse off after introducing a higher minimum wage. EDIT Can't resist anymore https://xkcd.com/552/ https://xkcd.com/552/

gizmo686 10y ago
"Correlation doesn't imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing 'look over there'." [0] [0] https://xkcd.com/552/ https://xkcd.com/552/ (alt text)
danarlow 10y ago
True, but it pretty definitively refutes the hypothesis that raising the minimum wage to $15 would dramatically increase unemployment in any reasonable setting. That is an important result. Now can have a discussion about the costs and benefits without the fear mongering. Also, here's the full article this blog is quoting: https://www.bloomberg.com/view/articles/2016-10-21/doomsayers-keep-getting-it-wrong-on-higher-minimum-wages https://www.bloomberg.com/view/articles/2016-10-21/doomsayer...
- icebraining 10y ago
  it pretty definitively refutes the hypothesis that raising the minimum wage to $15 would dramatically increase unemployment How so, if Seattle doesn't have a $15/h minimum wage? It'll only take effect in 2017, and even then, only for large employers that don't pay medical benefits. Under this law, the $15 wage will only be universal in 2021: http://murray.seattle.gov/minimumwage/ http://murray.seattle.gov/minimumwage/
jseliger 10y ago
Yes. Seattle also has an incredibly strong economy and (sadly) very tight housing policy: http://jakeseliger.com/2015/09/24/do-millennials-have-a-future-in-seattle-do-millennials-have-a-future-in-any-superstar-cities/ http://jakeseliger.com/2015/09/24/do-millennials-have-a-futu... that likely inhibits low-wage workers from living (or moving) there at all. In the meantime, many Texas cities: http://time.com/80005/why-texas-is-our-future http://time.com/80005/why-texas-is-our-future are building tons of affordable housing, even as superficially liberal cities like Seattle legislate it out of existence.
- NotSammyHagar 10y ago
  Seattle also has a limited land area, because of the ocean and lakes. There is only one way to get more housing, build up, or build outside the city limits.
  - schoen 10y ago
    Houseboats? Maybe more of those would come in handy here in San Francisco too!
    - Flammy 10y ago
      Seattle actually has some houseboats [1]. Not sure total numbers, but they are definitely a very small drop in the bucket compared to all other housing forms. Probably more impactful in terms of population numbers is the similar "apartment block over water" which popped up in Seattle a few decades back and are still around. ~5 stories and over water is a lot more people than a row of houseboats. [1] https://www.google.com/maps/place/Floating+Homes+Association/@47.6387207,-122.3293724,706m/data=!3m1!1e3!4m5!3m4!1s0x0:0xd59ce2356327fd0e!8m2!3d47.6407022!4d-122.3298387!6m1!1e1 https://www.google.com/maps/place/Floating+Homes+Association...
    - edwhitesell 10y ago
      If I'm not mistaken, houseboats in the Seattle area also have property tax tied to them as well. Instead of just a slip rental/lease. It been a few years since I heard that though, it may be incorrect or outdated.
tpeo 10y ago
You're right in pointing out that there might be unnacounted variables which might confound the impact of the minimum wage hike. But the majority of people running around saying that higher minimum wages cause unemployment do so on the basis of an overly simplified model (the standard partial equilibrium from Econ 101, where they're looking only at supply and demand curves for labor under complete information). Once you start weaking some conditions (using a search and matching framework instead of contracting under complete information) and adding confounding variables (like economy wide-effects or consumer credit restritions), not only your conclusions might change, but you might actually get a better model. Academics have some grasp of this and are much more subdued in their conclusions, specially when coming from pure theory. Pundits, however, are paid to be clueless. And I think this is the true insight of this submission.
mooman219 10y ago
I would like to point out a few things. First, "unemployment rate" is extremely subjective, with the Bureau of Labor listing 6 different possible measures that all fit the definition for "unemployment rate" [http://www.bls.gov/lau/stalt.htm http://www.bls.gov/lau/stalt.htm]. Not devaluing your results, but numbers reported online should be taken with a gain of salt. Second, I don't think it's completely fair to use this as precedent for cases elsewhere in the country. Seattle's mean income ($75,331) is well above both Washington State's ($64,129) and the US's rate ($55,775) [http://www.deptofnumbers.com/income/washington/seattle/ http://www.deptofnumbers.com/income/washington/seattle/]. The impact of adjusting minimum wage might be extremely understated.
- dlp211 10y ago
  > First, "unemployment rate" is extremely subjective No it isn't. At least not in this context or most contexts. While you are correct that there are multiple ways to measure unemployment, they are all objective measures. In addition to this, the colloquial usage of the "unemployment rate" refers to the U3 rate listed by the BLS.
firethief 10y ago
> The fall can be explained by an improving economy. The hypothesis of the article is that the economic benefit of increasing minimum wage is mediated by improvement of the economy due to increased spending. I'm not saying you're wrong, but that particular point doesn't seem to contradict the article's premise.
- sytse 10y ago
  The economy (and jobs) improved in the whole of the US so I think people can't claim it was caused by Seattle adopting the minimum wage. I'm not saying you or the article made that claim, just that it doesn't follow from this case.
Analemma_ 10y ago
Sigh. I have a dream when one day people learn to use that quote correctly. Correlation does not imply causation, but a lack of correlation does suggest (not prove, but suggest) a lack of causation. In this case, the relevant claim is the claim by minimum wage opponents that the unemployment rate and the minimum wage are positively correlated. Seattle's situation does NOT imply that a $15/hr minimum wage decreases unemployment (if someone were to make that claim, then your quip would apply), but it DOES suggest that a $15/hr minimum wage does not increase unemployment; this distinction is subtle but vital.
- geomark 10y ago
  "Correlation does not imply causation, but a lack of correlation does suggest (not prove, but suggest) a lack of causation." I don't think it's any stronger one way or the other. Correlation does not imply causation; it may suggest it, but there may be confounders. And lack of correlation does not imply lack of causation, because there may be confounders that mask results that would suggest causation. Showing causation is hard and usually takes careful design of experiment with very good controls to eliminate any potential confounders, something that is not present in retrospective studies.
hristov 10y ago
Correlation is not always causation but it can be and it is usually very strong evidence against the presence of opposite forces of causation. If higher minimum wages actually cause unemployment (as minimum wage opponents often argue) then how do you explain Seattle.

by sytse 10y ago

gizmo686 10y ago

"Correlation doesn't imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing 'look over there'." [0] [0] https://xkcd.com/552/ https://xkcd.com/552/ (alt text)

danarlow 10y ago

True, but it pretty definitively refutes the hypothesis that raising the minimum wage to $15 would dramatically increase unemployment in any reasonable setting. That is an important result. Now can have a discussion about the costs and benefits without the fear mongering. Also, here's the full article this blog is quoting: https://www.bloomberg.com/view/articles/2016-10-21/doomsayers-keep-getting-it-wrong-on-higher-minimum-wages https://www.bloomberg.com/view/articles/2016-10-21/doomsayer...

icebraining 10y ago
it pretty definitively refutes the hypothesis that raising the minimum wage to $15 would dramatically increase unemployment How so, if Seattle doesn't have a $15/h minimum wage? It'll only take effect in 2017, and even then, only for large employers that don't pay medical benefits. Under this law, the $15 wage will only be universal in 2021: http://murray.seattle.gov/minimumwage/ http://murray.seattle.gov/minimumwage/

jseliger 10y ago

Yes. Seattle also has an incredibly strong economy and (sadly) very tight housing policy: http://jakeseliger.com/2015/09/24/do-millennials-have-a-future-in-seattle-do-millennials-have-a-future-in-any-superstar-cities/ http://jakeseliger.com/2015/09/24/do-millennials-have-a-futu... that likely inhibits low-wage workers from living (or moving) there at all. In the meantime, many Texas cities: http://time.com/80005/why-texas-is-our-future http://time.com/80005/why-texas-is-our-future are building tons of affordable housing, even as superficially liberal cities like Seattle legislate it out of existence.

NotSammyHagar 10y ago
Seattle also has a limited land area, because of the ocean and lakes. There is only one way to get more housing, build up, or build outside the city limits.
- schoen 10y ago
  Houseboats? Maybe more of those would come in handy here in San Francisco too!
  - Flammy 10y ago
    Seattle actually has some houseboats [1]. Not sure total numbers, but they are definitely a very small drop in the bucket compared to all other housing forms. Probably more impactful in terms of population numbers is the similar "apartment block over water" which popped up in Seattle a few decades back and are still around. ~5 stories and over water is a lot more people than a row of houseboats. [1] https://www.google.com/maps/place/Floating+Homes+Association/@47.6387207,-122.3293724,706m/data=!3m1!1e3!4m5!3m4!1s0x0:0xd59ce2356327fd0e!8m2!3d47.6407022!4d-122.3298387!6m1!1e1 https://www.google.com/maps/place/Floating+Homes+Association...
  - edwhitesell 10y ago
    If I'm not mistaken, houseboats in the Seattle area also have property tax tied to them as well. Instead of just a slip rental/lease. It been a few years since I heard that though, it may be incorrect or outdated.

tpeo 10y ago

You're right in pointing out that there might be unnacounted variables which might confound the impact of the minimum wage hike. But the majority of people running around saying that higher minimum wages cause unemployment do so on the basis of an overly simplified model (the standard partial equilibrium from Econ 101, where they're looking only at supply and demand curves for labor under complete information). Once you start weaking some conditions (using a search and matching framework instead of contracting under complete information) and adding confounding variables (like economy wide-effects or consumer credit restritions), not only your conclusions might change, but you might actually get a better model. Academics have some grasp of this and are much more subdued in their conclusions, specially when coming from pure theory. Pundits, however, are paid to be clueless. And I think this is the true insight of this submission.

mooman219 10y ago

I would like to point out a few things. First, "unemployment rate" is extremely subjective, with the Bureau of Labor listing 6 different possible measures that all fit the definition for "unemployment rate" [http://www.bls.gov/lau/stalt.htm http://www.bls.gov/lau/stalt.htm]. Not devaluing your results, but numbers reported online should be taken with a gain of salt. Second, I don't think it's completely fair to use this as precedent for cases elsewhere in the country. Seattle's mean income ($75,331) is well above both Washington State's ($64,129) and the US's rate ($55,775) [http://www.deptofnumbers.com/income/washington/seattle/ http://www.deptofnumbers.com/income/washington/seattle/]. The impact of adjusting minimum wage might be extremely understated.

dlp211 10y ago
> First, "unemployment rate" is extremely subjective No it isn't. At least not in this context or most contexts. While you are correct that there are multiple ways to measure unemployment, they are all objective measures. In addition to this, the colloquial usage of the "unemployment rate" refers to the U3 rate listed by the BLS.

firethief 10y ago

> The fall can be explained by an improving economy. The hypothesis of the article is that the economic benefit of increasing minimum wage is mediated by improvement of the economy due to increased spending. I'm not saying you're wrong, but that particular point doesn't seem to contradict the article's premise.

sytse 10y ago
The economy (and jobs) improved in the whole of the US so I think people can't claim it was caused by Seattle adopting the minimum wage. I'm not saying you or the article made that claim, just that it doesn't follow from this case.

Analemma_ 10y ago

Sigh. I have a dream when one day people learn to use that quote correctly. Correlation does not imply causation, but a lack of correlation does suggest (not prove, but suggest) a lack of causation. In this case, the relevant claim is the claim by minimum wage opponents that the unemployment rate and the minimum wage are positively correlated. Seattle's situation does NOT imply that a $15/hr minimum wage decreases unemployment (if someone were to make that claim, then your quip would apply), but it DOES suggest that a $15/hr minimum wage does not increase unemployment; this distinction is subtle but vital.

geomark 10y ago
"Correlation does not imply causation, but a lack of correlation does suggest (not prove, but suggest) a lack of causation." I don't think it's any stronger one way or the other. Correlation does not imply causation; it may suggest it, but there may be confounders. And lack of correlation does not imply lack of causation, because there may be confounders that mask results that would suggest causation. Showing causation is hard and usually takes careful design of experiment with very good controls to eliminate any potential confounders, something that is not present in retrospective studies.

hristov 10y ago

Correlation is not always causation but it can be and it is usually very strong evidence against the presence of opposite forces of causation. If higher minimum wages actually cause unemployment (as minimum wage opponents often argue) then how do you explain Seattle.