Our Own Troll “Joe” Shows How to Pick Cherries
In my post about part-time versus full-time employment, frequent commenter and critic Joe wrote:
It’s a lot of fun to play the Obamacare vs the Labor Force game. The best one is looking at private sector employment before and after Obamacare was signed into law on March 2010.
Private Payroll Employment (thousands)
March 1999 107938
March 2010 106961
March 2013 113454So in the 11 years prior to the passage of Obamacare, fewer than zero jobs were added to the private sector. In the 3 years following the passage of Obamacare, nearly 7 million jobs were added to the private sector.
It’s quite the job killer.
Now two points of clarification:
==> In my post, I wasn’t saying ObamaCare had caused the change in PT/FT employment. That was actually my point, that it had moved earlier and so (at best) could be blamed on more general “regime uncertainty.”
==> It’s conceivable Joe is just being ironic, and wants to show how one could be extremely misleading with statistics if one wanted. (He does that a lot in the comments here…)
Anyway, I thought this was a good opportunity to show how one can indeed mislead with statistics. I didn’t check the exact numbers, but I have no reason to doubt Joe’s claims above. Yet look at the chart:
Last thing: I know conventional wisdom says, “Don’t feed a troll,” but my plug-in analog clock says it’s before midnight so we should be OK.
I call to Eric’s attention the missing lower part of the graph. Yet Bob has clearly copied a graphic from FRED, a reputable source. Once again, there is no one size fits all way to show numbers. It’s what people do with the graphic or the context that can make some msileading.
Ken B: I get it. My point isn’t Bob’s graph, it is the original graph in the original post. It’s definitely a lies, damn lies graph.
Plot LNS12500000 (Normally Full Time) and LNS12032194 (Part Time for all Reasons) using FRED on one graph, same axis, and you’ll see my point. Part Time is really flat since at least 1982 (it goes back further, but I only went that far back on my graph).
I can’t figure out how to post the graph here or I’d do it… although this might do it.
The original graph is as % of all employment. I think it tells you something, albeit something different from a graph of employment rates. If numbers of part time jobs stays the same an total employment falls then we would get this sort of graph. The “problem” with it is redundancy – the two curves are required to mirror each other, so in effect only plotting one of them would convey as much information.
Ahhhhh. I didn’t see that. If he’s talking as % of employed, rather than % of employable then yes you are right. The mirroring is by definition. It doesn’t display evidence of a corelation.
Right, the mirroring doesn’t add anything.
Besides the cherry picking of dates, there is another argument against the “Obama Care is not a job killer” claim by our friend Joe: Simply put, we don’t know yet if Obama Care will be a net job creator or job killer (As if that is the goal). The law doesn’t go into effect until January 2014! There is no way you can draw any Obama Care”impact” conclusion from Bob’s helpful chart.
When the law passed initially, nobody (including the guys and gals voting for it) had actually read the bill. In fact, law makers were encouraged by leadership (Nancy P.) to vote for it, so they could find out what was in it (…).
All we (the unsuspecting public) were told was:
– Nothing will change if you already have coverage. You are “good to go”.
Based on that, most of us went back to what they were doing before (i.e. living life), thinking this law would have zero impact on us (While nefarious forces started writing the draconian regulations). It wasn’t until recently that the enormity of the actual regulations was understood. This process is still ongoing.
Now that the changes are becoming more widely understood, individuals, CFO’s and small business owners are reacting to them. And although it is hard to make predictions (And “us Austrians” are not in that business), my money is on them following basic rules of economics, and that we’ll see a lot more “49’ers” (Companies with 49 employees or less) and “29’ers” (Employees working 29 hours or less), an more individuals who will roll the dice and pay the fine (strike that, “tax”).
Hey Murphy, you might like this. After the Fed announced it will reduce its QE from $85bb a month down to $75bb a month, the stock markets all jumped and closed the day at relative highs.
Email Scott Sumner and tell him that the stock market doesn’t like QE, hahaha
“Never reason from a price change, Bob.”
“…but only when the stock market reacts differently from what MM predicts.”
Just guessing: They had priced in a taper to $65b?
This one’s good too:
Krugman Dec 2013:
http://www.nytimes.com/2013/12/02/opinion/krugman-better-pay-now.html?hp&rref=opinion&_r=4&
“So can anything be done to help these workers, many of whom depend on food stamps — if they can get them — to feed their families, and who depend on Medicaid — again, if they can get it — to provide essential health care? Yes. We can preserve and expand food stamps, not slash the program the way Republicans want. We can make health reform work, despite right-wing efforts to undermine the program.
And we can raise the minimum wage.”
Krugman 1998:
http://www.thefreelibrary.com/Living+Wage:+What+It+Is+and+Why+We+Need+It.-a021103427
“So what are the effects of increasing minimum wages? Any Econ 101 student can tell you the answer: The higher wage reduces the quantity of labor demanded, and hence leads to unemployment.”
Good analysis here:
http://www.huffingtonpost.com/ben-powell/krugman-minimum-wage_b_4428174.html
If you insist that correlation implies causation, then you can prove ANYTHING.
Besides relying on that pathetic fallacy(not every correlation is causally related), he ignores persons employed as a percentage of population.
Government spending is….. like 1/3 higher now?
So let’s use that correlation and imply causation
http://data.bls.gov/timeseries/LNS11300000
“It’s quite the job killer.”
The correct thing to say is that employment could be HIGHER without xyz government policy. If there is YoY economic growth (with huge government), this doesn’t mean government caused growth. B/C growth may have been even higher without that huge government. That means that the bloated government is a DRAG on economic growth. I.E. 3 % yoy growth with monster gov’t versus 5% YoY without.
49% of the whole population are receiving government benefits/assistance.
Typical Keynesian response: “Our standard of living is higher than when there was 0% dependent on government assistance.”
Or
Pelosi: Unemployment checks create jobs.