Governing through Reason
Unemployment and Reëlection
Statistician Nate Silver crunched the numbers, and found that there is no particular correlation between the change in unemployment over a President’s entire first term and his chances of reëlection to a second term. Thus, President Obama’s reëlection chances do not appear as daunting as today’s 9.1% unemployment rate suggest.
The problem with Mr. Silver’s analysis is that voters do not survey the net change over an entire term and award their vote accordingly. Rather, they consider where unemployment and the rest of the economy is heading at the time they cast their ballot. Thus, the better way of measuring the effect of unemployment on reëlection is the change from the prior high or low in unemployment during the middle of the President’s term to the unemployment on election day.
For example, among the post-WWII Presidents seeking reëlection when unemployment was above 7% on election day, Reagan won because unemployment fell by about three percent from its prior high in 1983, while Carter and Bush 41 lost because unemployment increased by about 2% from lows before the election.
The other post WWII Presidents who won reëlection to a second term all enjoyed nearly full employment on election day falling off a prior high.
Obama is unique among post WWII Presidents in that his election day unemployment rate will likely be between 8% to 9%, probably closer to 9%. The only prior 20th century President to seek a second term with unemployment so high was FDR in 1936.
FDR won because unemployment fell several percent before the 1936 election. In Obama’s case, unemployment has not actually fallen at all. The nominal drop from 10% to 9% was primarily because a couple million unemployed simply stopped looking for work and were no longer counted in the figure. Even that faux drop appears to be reversing as the early stages of stagflation appear to be throwing more folks out of work.
In sum, there are no data points for a President like Obama whose real unemployment rate flatlined at a high rate for the last three years of his first term. Even so, I doubt any President seeking reëlection since WWII would have wanted to trade places and run on Obama’s economy.
Related articles
- Unemployment rate does not necessarily predict re-election chances (dailykos.com)
- Will Jobs Numbers Sink Obama in 2012? (thedailybeast.com)
- President Obama And Unemployment (alan.com)
- Economic news is bad for Obama’s reëlection bid — Washington Post (news.google.com)
- Troubling Signs for Obama’s Re-Election (politicalwire.com)
- Economic Recovery? (kajunman.wordpress.com)
-
You take correlations and place them in the context of all relevant facts.
The facts are:
1) Who ran.
2) Who won.You have furnished no mechanism to support your contention, either. You have insisted correlation is sufficient. My theory is as convincing as yours, and has more hard evidence (i.e., more datapoints) to support it.
My own suspicion is that economic conditions (or, more accurately, the public perceptions of economic conditions) influence (but do not determine) voting patterns. Unemployment (the actual rate, the direction of movement, the perception of rate and movement) is one of many factors that influences the public’s perception of economic conditions.
It is not causative, deterministic, or sufficient. Since you have not postulated any mechanism (just as you did not with your correlation of speeches and the movement of the DOW) there is no evidence whatever that it is anything more than a distantly-related side-effect, if even that. Further, since you have nothing but a cherry-picked and completely subjective (therefore random) “highest peak / lowest valley” to go on, your theory isn’t even testable.
Mine is specific, testable, repeatable, and quantitative. Yours is none of those things — it is only propagandistic.
-
#203 written by mclever 1 year ago
*sigh*
“Proof” has different meanings to people in different fields, as evidenced by the back and forth between Bart and Michael here.
To Bart (and most legally-minded folks), “proof” implies the legal definition of “evidence beyond a reasonable doubt,” which obviously allows for “common sense” extrapolations based on inferences and apparent correlation. “It’s true because it looks true based on everything we can see.”
To Michael (and most scientifically-minded folks), “proof” means it has to pass the standard of falsifiability and testability to confirm not just an apparent correlation but a definitive causation. This is a higher standard than just assuming something is true because it looks true. Nine coin-flips coming up in a row as heads doesn’t prove the coin is weighted, for example. Barring being able to measure the coin, we’d need a much larger sample of coin-flips to make that determination.
The correlation between unemployment numbers and re-election chances has some predictive value, but it isn’t the “proven cold hard fact” that some make it out to be. Just because something hasn’t happened yet doesn’t mean it never will. Just because the team getting out to a 3–0 start in the Stanley Cup Playoffs has won 24 out of 25 times, that doesn’t mean that a 3–0 start is a guaranteed win. Just because Djokovich won 43 matches in a row to start this year, that didn’t guarantee he’d win match #44…
The history since WWII does make a compelling case for the difficulty of being re-elected with high (or trending upward) unemployment, but it does not constitute proof that someone cannot be re-elected in a poor employment environment.
-
Mclever:
To Bart (and most legally-minded folks), “proof” im plies the le gal de f i n i tion of “ev i dence be yond a rea son able doubt,” which ob vi ously al lows for “com mon sense” ex trap o la tions based on in ferences and ap par ent cor re la tion.
To clarify, proof in law is simply the application of relevant evidence to an issue. Beyond a reasonable doubt is the standard of proof for a criminal conviction. There are other lesser standards for other issues.
I agree that Michael is coming from a different mathematical perspective. I also understand his standard of proof, but Michael refuses to admit to other standards and instead accuses me of acting in bad faith. This does not advance the discussion.
While statistical reliability is a useful standard, the vast majority of the world correctly and successfully applies evidence to make decisions without applying or often even knowing about statistical reliability.
The his tory since WWII does make a com pelling case for the dif fi culty of be ing re-elected with high (or trend ing up ward) un em ploy ment, but it does not con sti tute proof that some one can not be re-elected in a poor em ploy ment environment.
I agree. My article stated that Obama is in a unique reëlection situation without post WWII precedent in that he has basically flatlined high unemployment. There are no 2–3% changes in the unemployment rate as there were with other presidents with high unemployment like Carter, Reagan and Bush41.
We have a bad economy without trends to suggest to voters whether it is getting better or worse. I suspect that a President with a bad economy which is not getting better is in trouble. We will have to wait to see how that plays out in 2012.
-
#205 written by shortchain 1 year ago
-
#206 written by mclever 1 year ago
I think the point is, Bart, that you have a tendency to state opinion as if it were fact, and similarly to incorrectly equate implied evidence with “proof” of your opinion. We all have that tendency to some extent, so we should all be careful about overstating opinions.
A logical correlation may convince a typical jury, but that doesn’t mean it actually constitutes “proof.” Given the academic attitudes of many folks here, this “jury” is a little more demanding on its standards of proof, and it’s unlikely to accept conjecture as fact without proof beyond merely a compelling list of coincidences.
If you present conjecture as fact, expect someone here to challenge you for clarification and proof of your assertions. Otherwise, you may be better served by making sure you phrase things in the form of “It seems to me” or “I think that” rather than the more assertive “It is so.”
-
Bart,
Where precisely did I take and decline data “on faith” concerning your false claim that I accused BLS of “cooking their books?”.
Wrong antecedent. You take and decline employment data on faith.
You have admitted that you believe unemployment combined with the economy has a plurality effect on the outcome of elections.
Yes. Because I do.
A plurality effect on the outcome of a election decides the election unless all other factors align for a President seeking reelection, which has never happened during the post WWII period we are studying.
No. Plurality means that it has more of an impact than any other single item. One only needs a significant contingent of the other factors to align for a President seeking reëlection. I agree that it hasn’t happened in the Post WWII period. At the moment, it looks plausible for it to happen this time. If the Republicans can get a 2012 candidate going who generates widespread enthusiasm, things look bleak for Obama. They haven’t done it yet.
-
#209 written by mclever 1 year ago
BTW, Bart:
“Proof beyond a reasonable doubt” is actually a LESS strict standard.
“Proof beyond all possible doubt” is what “proof” means in the strictest sense. Scientists are very particular about what something means to be “proven.”
You may have presented a compelling case, but that doesn’t mean it equals proof in an absolute sense. If “proof beyond a reasonable doubt” were the strictest of standards, then there would never be a false conviction, because the proof would be incontrovertible.
-
Bart,
Michael refuses to admit to other standards
I admit that they exist. However, conflating colliding nomenclature within a single issue discussion succeeds only in increasing confusion. When the discussion shifted from the qualitative argument you initially put forth to an attempted quantitative confirmation, continuing to use qualitative nomenclature results in inappropriate implications.
In other words, when we start talking statistics, words like “proof” no longer have any meaning.
and instead accuses me of acting in bad faith.
My bad-faith accusation applies to your apparent belief that BLS does not accurately report U3. But here’s your chance to set the record straight: do you believe that BLS is accurately reporting U3, as described by the UN definition?
the vast majority of the world correctly and successfully applies evidence to make decisions without applying or often even knowing about statistical reliability
Not to go too far off on a tangent, but while they do apply evidence to make decisions without applying or even knowing about statistical reliability, they do so incorrectly and unsuccessfully far more often than most people realize. Confirmation bias causes them to believe that they are more successful at it than they really are.
-
#211 written by Max aka Birdpilot 1 year ago
mc,
Bart, a lawyer, is attempting to use statistical terminology while ignoring statistical requirements. He is attempting to explain stochastically independent events, elections, as though they are dependent and, worse yet, attempting to so simplify as to assign a single determining factor to the outcome.
He wants to use the courtroom test, simply, whether or not a jury can be convinced of an outcome, upon being shown “evidence”. A statistician has a litany of terminology to describe what and how to reach a conclusion to a certain degree of confidence.
Which is why dc can demonstrate the fallacy of Bart’s conclusion in search of a premise by showing a higher correlation simply by looking at the last letter of the last name.
Bart will never admit his attempt at bastardization. It’s why I refuse to engage in his silliness.
-
Mclever:
…you may be bet ter served by mak ing sure you phrase things in the form of “It seems to me” or “I think that” rather than the more as sertive “It is so.”
Legal analysis offers two different approaches — a neutral approach that judges are supposed to take weighing the pros and cons of an issue and a persuasive approach where an advocate attorney offers his or her proof and argues for a proposition. I take the latter approach because its is consistent with my nature and my training in the military and in trial law.
The surest way to lose a persuasive argument is to act as if you are not sure of your facts and start hedging with qualifiers like “It seems to me.” If you do not believe in what you are saying, why should anyone else. Indeed, hedging makes for a lousy neutral approach as well and good court opinions rarely engage in it.
Now a persuasive argument is only opinion when unsupported by evidence and if it does not rule out contrary evidence. This is why I will not make a persuasive assertion without evidence and challenge others to come up with contrary evidence with the objective of ruling it out to prove my proposition.
My ideological opposites may not like my proposition or my evidence, but there is always evidence and frequent links that evidence. The old 538 Refugees site kept throwing me into moderation limbo because I included too many links.
I carefully consider the evidence pro and con for a position before posting to ensure that my position is well supported and my ideological opposites here will not be able to come up with contrary evidence which I cannot address. Before writing this article, I had carefully considered Nate’s analyses and simply could not square his analysis with the facts as they existed. You never go to trial without knowing the answers to all the questions ahead of time.
When there is no substantive evidence — as with the Obama reëlection chances in a situation with flatlining high unemployment — I start using the preferred legal hedging terms of “I suspect” and the like. But where there is persuasive evidence, I will not hedge.
If you want mush, read another commentator.
-
Mclever:
“Proof be yond a rea son able doubt” is ac tu ally a LESS strict standard. “Proof be yond all pos si ble doubt” is what “proof” means in the strictest sense. Sci en tists are very par tic u lar about what some thing means to be “proven.”
Because humans are always fallible, “Proof be yond all pos si ble doubt” is simply impossible.
Scientists are human beings too and are not capable of “Proof be yond all pos si ble doubt.” Indeed, as fallible human beings subject to the same influences of fame and money as anyone else, they often employ rather loose standards of proof.
I do not subscribe to this worship of credentialing which is endemic among progressives. On every issue under the sun, you can find scientists who are willing to testify for either side for money. I have cross examined and discredited too many “experts” at trial to give their evidence any particular credence simply because they have a credential. Indeed, jury instructions routinely warn jurors not to do so.
-
#214 written by shortchain 1 year ago
-
Bart,
The surest way to lose a persuasive argument is to act as if you are not sure of your facts and start hedging with qualifiers like “It seems to me.”
That works in a courtroom, because the jurors are not able to challenge your assertions. Instead, the rules of debate are very tightly constrained, which causes any squishiness to be perceived as weakness.
In a debate, however, the surest way to lose is to pretend that qualifiers are unnecessary, because the situation is clear. In such a situation, when a qualifier is shown to be necessary, the rest of the argument is thrown into question. Retrenching only serves to reinforce the doubt.
If I were in a courtroom, I wouldn’t be including the qualifiers. I would be arguing as you do, using as many techniques as possible to obfuscate and dissemble evidence to the contrary of my position. The rules in a courtroom would work to my advantage if I take that approach…in a courtroom. But (looking around the site) this is not a courtroom. I tailor my approach to the venue. You would be wise to do the same.
-
Bart,
I do not subscribe to this worship of credentialing which is endemic among progressives.
And yet I presume you still visit doctors, rather than treating yourself for all illnesses. And I presume that you don’t advise your clients to forgo representation by legal council, despite it being perfectly legal to do so.
Why is that?
-
MW:
BD: A plu ral ity ef fect on the out come of a elec tion de cides the elec tion un less all other fac tors align for a Pres i dent seek ing re elec tion, which has never hap pened dur ing the post WWII pe riod we are studying.
No. Plu ral ity means that it has more of an im pact than any other sin gle item. One only needs a sig nif i cant con tin gent of the other fac tors to align for a Pres i dent seek ing re elec tion. I agree that it hasn’t hap pened in the Post WWII pe riod. At the mo ment, it looks plau si ble for it to hap pen this time. If the Re pub li cans can get a 2012 can di date go ing who gen er ates wide spread en thu siasm, things look bleak for Obama. They haven’t done it yet.
OK, now support this opinion with actual facts. What combination of factors are aligning for Mr. Obama that would outweigh unemployment and the economy and lead to his reëlection? And what evidence supports the proposition?
Your and others’ suggestion that Mr. Obama’s megawatt charisma in comparison to any of the dull Elephants running for his job will overcome his poor job approval in addressing a flatlining economy requires proof.
BD: …and in stead ac cuses me of act ing in bad faith.
My bad-faith ac cu sa tion ap plies to your ap par ent be lief that BLS does not ac cu rately re port U3. But here’s your chance to set the record straight: do you be lieve that BLS is ac cu rately re port ing U3, as de scribed by the UN definition?
I never stated nor implied that the BLS was not following its rules for compiling U3 and do not do so now. I used the U3 data in my article. Claiming that BLS is cooking my evidence hardly helps my argument. I don’t anticipate that you will have the class to admit this. I will be satisfied if you stop knocking down this straw man.
-
MW:
BD: The surest way to lose a per sua sive ar gu ment is to act as if you are not sure of your facts and start hedg ing with qual i fiers like “It seems to me.”
That works in a court room, be cause the ju rors are not able to chal lenge your as ser tions. In stead, the rules of de bate are very tightly con strained, which causes any squishi ness to be per ceived as weakness.
:::chuckle:::
We have an adversarial legal system. There will be one or more attorneys on the other side whose job is to test your evidence.More recently, CO and other states allow jurors to ask questions of witnesses which we also treat in closing arguments as questions for the attorneys.
Finally, you can get almost any relevant non-hearsay evidence into a court of law.
-
#219 written by Max aka Birdpilot 1 year ago
-
MW:
BD: I do not sub scribe to this wor ship of cre den tial ing which is en demic among progressives.
And yet I pre sume you still visit doc tors, rather than treat ing your self for all ill nesses. And I presume that you don’t ad vise your clients to forgo rep re sen ta tion by le gal coun cil, de spite it be ing per fectly le gal to do so.
Just because you go to a doctor or lawyer for their services hardly means that you accept without question their qualifications or what they tell you, or decline to shop around simply because they are credentialed.
I pepper my doctor with questions until I am satisfied. I would hardly be shy about requesting a second opinion and have advised my relatives to do so on occasion.
We attorneys are also nothing special.
When I meet with a prospective client, I spend considerable time reviewing their options and providing pros and cons. And yes I often advise prospective clients they often have the option of foregoing expensive legal representation when other alternatives exist to resolve their problems. I do this for free. The information and candor generally impresses prospective clients. Even if they take a non-attorney option, I often get them as clients later for actual legal problems or get referrals of their friends and family.
Most attorneys are not nearly so open, so I advise my friends and family to question prospective attorneys and shop around if their answers do not sound right.
NEVER accept what an attorney, doctor or any other professional says at face value just because they have a nice yellow pages ad, fancy office or lots of certificates on the wall. As an intelligent human being, you are perfectly capable of asking questions, gathering information and making your own decisions on anything. Exercise your free will.
-
-
OK, now support this opinion with actual facts. What combination of factors are aligning for Mr. Obama that would outweigh unemployment and the economy and lead to his reelection?
No problem. Imagine, for the moment, that the Republican candidate was Osama bin Laden, and no significant third-party candidate showed up. Would you disagree that Obama would win in a landslide? Yes, this example is extreme, but it’s there to prove a point: a truly horrible Republican candidate would lose to Obama even in a dreadful economy.
Obviously, the Republicans won’t be putting a dead terrorist on the ballot. But they can put someone on who is a real turnoff to voters. The remaining question is whether the “badness” of the Republican candidate would outweigh the “badness” of the economy.
Your and others’ suggestion that Mr. Obama’s megawatt charisma…
Not my suggestion. I’ll leave others to defend his charisma as being a significant contributing factor to the upcoming election.
…in comparison to any of the dull Elephants running for his job…
I’d hardly call Palin, Bachmann, Trump, or Gingrich “dull”. They are repulsive to significant segments of the voting population. That’s the problem, not the dullness.
…requires proof.
In the case of Palin, for instance, the number of poll respondents who claimed that they absolutely would not vote for Palin was so overwhelming that it doesn’t look like it would matter that she’d run against an incumbent with a bad economy.
I never stated nor implied that the BLS was not following its rules for compiling U3 and do not do so now.
Good enough for me on that front. Do you further concur that BLS reports U3 because that’s the apples-to-apples comparison that has been made for decades, and not because it looks better than having a top-line report of U4, U5, or U6?
I don’t anticipate that you will have the class to admit this.
To admit what?
-
Bart,
Just because you go to a doctor or lawyer for their services hardly means that you accept without question their qualifications or what they tell you, or decline to shop around simply because they are credentialed.
Which is why it’s perfectly reasonable to look at several credentialed people’s conclusions, rather than merely one. But it sure doesn’t mean that it’s a good idea to use Sarah Palin as a reference when looking at US history. It means looking at the viewpoints of several people who specialize in US history.
But I’m sure not going to go to an auto mechanic for heart surgery. And I’m not going to look to a think tank for a realistic perspective on global climate.
-
Bloomberg just issued the usual Dem media biased poll of non-voting adults with a Dem+5 ratio and about 6% fewer whites than actually vote. Even this gamed poll did not help President Obama’s reëlection prospects:
45% think that unemployment and jobs is the #1 issues, followed by 17% and 15% for government spending and deficit, respectively. The Dem issue of health care came in 4th with 10%.
76% properly view the economy a unchanging or getting worse.
55% think the proper approaches to return economic growth are to cut spending and taxes.
57% disapprove of Obama’s job performance with the economy and jobs.
40% of “likely voters” say they will definitely vote against Obama in 2012 vs. 31% who will definitely vote for him.
61% agree that “in 2012, President Obama will have had his chance — if the economy isn’t substantially better by election day, it will be hard to vote for him.“
-
-
#226 written by Max aka Birdpilot 1 year ago
-
45% think that unemployment and jobs is the #1 issues
And it seems from Battleground that Obama has the lead on this issue.
followed by 17% and 15% for government spending and deficit, respectively
Which are the issues on which the Republicans have the lead.
The Dem issue of health care came in 4th with 10%.
Good thing Obama has the #1 issue, too. Advantage, Obama.
76% properly view the economy a unchanging or getting worse.
Properly? Leaving the editorialization aside, their view is what matters in the polling booth.
55% think the proper approaches to return economic growth are to cut spending and taxes.
That’s not what the poll asked. That’s what the poll should have asked, but it didn’t.
57% disapprove of Obama’s job performance with the economy and jobs.
And yet, of those polled on Obama’s economic vision for the future versus the Republicans, Obama was preferred. So, if you combine the two, you get the following message:
“Obama hasn’t been good on the economy and jobs…and the Republicans are even worse.“
Now that’s a campaign slogan for T-Paw!And, of those polled, more were worried about a Republican-run House, Senate, and Presidency than were worried about maintaining the status-quo.
61% agree that “in 2012, President Obama will have had his chance — if the economy isn’t substantially better by election day, it will be hard to vote for him.“
Of course. At this point, I think the winner in 2012 would be “None of these” if the national election were like the Nevada elections. But it’s not. I’ve said this for a while…a good Republican candidate should win this without even breathing hard. And yet…
60% agree that “To win the nomination, the Republican nominee will have moved so far to the right on fiscal and social issues it will be very hard to vote for that candidate” -
On the topic of unemployment and reëlection, Pew has Obama’s numbers back to and a bit worse than where they were before the 2010 election.
http://people-press.org/files/legacy-questionnaires/Econ%20topline%20for%20release.pdf
The only folks sticking with Obama are the Dems. His numbers are way underwater with Indis and Reps and in every region apart from the NE.
Party like its 1979!
-
You must be logged in to post a comment.
- Comment Feed for this Post
dc:
You take correlations and place them in the context of all relevant facts.
There are no other facts which indicate that the last letter of an incumbent’s name determines reëlection, thus we have a coincidence.
The proposition that unemployment trends determine presidential reëlection post-WWII has a perfect record and is confirmed by the observations of nearly every historian, political professional and lay person. On the other side of the ledger, no one here including Michael seriously proposes and offers evidence that there is another independent factor which is more important or even comes close in importance to determining reëlection.
Nate and Michael are wrapped up in their inability to mathematically prove this proposition. I wish them luck in the future. However, this inability does not discount the evidence on the table now.