Search






Jeff's Amazon.com Wish List

Archive Calendar

November 2024
M T W T F S S
 123
45678910
11121314151617
18192021222324
252627282930  

Archives

The Stats on Chrysler Dealership Closings [Dan Collins]

A couple of days ago, I asked for a statistician to analyze the data regarding the closings of Chrysler dealerships and the apparent relation to donations to Republicans. I’m not going to say that the following is rigorous, but it does at least give one a ballpark figure:

This work builds upon the research done by numerous parties, most notably Joey Smith. It is a follow-up to my original post, entitled “Did anti-Obama campaign contributions dictate which Chrysler dealers were shuttered?” The odds that these closings occurred without partisan bias are roughly equivalent to the odds that Jean Claude Van-Damme will grab a Best Supporting Actor Oscar next year for a remake of Terms of Endearment.

How did the U.S. government’s “car czar” decide which Chrysler dealers to close and which would remain open? No one appears to know, not even the President of Chrysler:

…Lawyer Leonard Bellavia, of Bellavia Gentile & Associates, who represents some of the terminated dealers, said he deposed Chrysler President Jim Press on Tuesday and came away with the impression that Press did not support the plan…

“It became clear to us that Chrysler does not see the wisdom of terminating 25 percent of its dealers,” Bellavia said. “It really wasn’t Chrysler’s decision. They are under enormous pressure from the President’s automotive task force.”

Follow the evidence trail, below, and judge for yourself.

Dealers on the closing list donated millions to Republicans, $200 for Obama
The initial pass at the list of shuttered dealers showed they had donated, in the aggregate, millions to Republican candidates and PACs and a total of $200 to Barack Obama.

In fact, I have thus far found only a single Obama donor ($200 from Jeffrey Hunter of Waco, Texas) on the closing list.

Another review of all 789 closing dealerships, by WND, found $450,000 donated to GOP presidential candidates; $7,970 to Sen. Hillary Clinton; $2,200 to John Edwards and $450 to Barack Obama.

Now, and this is important, Chrysler claimed that its formula for determining whether a dealership should close or not included “sales volume, customer service scores, local market share and average household income in the immediate area.”

Dealer Jim Anderer told Fox News’ Neil Cavuto he can’t comprehend how his dealership can be among those killed: he stated that his sales volume ranking is in the top 2 percent of all dealers.

*******

All other factors being equal, what are the odds that RLJ’s dealerships would remain open while all other area dealerships would be shuttered? The approximate odds of such an occurrence can be calculated. 789 of the Chrysler’s dealerships are closing, which represents 25% of the total (according to MSNBC).

Recall that Chrysler claimed that its formula for determining whether a dealership should close or not included “sales volume, customer service scores, local market share and average household income in the immediate area.”

Thus, the odds that any, randomly selected, single dealership would remain open is roughly 75%. The odds that a single dealership would close is roughly 25%.

In the Bentonville, AR territory, the odds that RLJ would remain while its competition gets axed is .75 * .25 * .25 = .046875 (4.6%).

In Huntsville, AL, the odds are .75 * .25 * .25 = .046875.

In Branson, MO, .75 * .25 *.25 *.25 = .01171875.

In Lee’s Summit, MO, .75 * .25 *.25 *.25 *.25 = .0029296875.

In Shreveport, LA, .75 * .25 *.75 = .140625.

What are the odds of all of these RLJ dealerships remaining open while their competitors are wiped out? Maybe 1/10,000,000 of 1%. Yes, that’s one ten-millionth of one percent.

Approximately the odds that I’ll win American Idol. Or that you’ll land two frisbees, simultaneously, on each of Barack Obama’s teleprompters during one of his televised speeches.

Hello, mainstream media: anyone listening? How about you, class-action lawyers?

So, it appears that the Obama administration has been able to use taxpayer monies to expropriate people’s properties on the basis of political concessions. What do you call that?

Little Miss Attila’s been doing yeoperson’s work on this matter. Keep scrolling.

144 Replies to “The Stats on Chrysler Dealership Closings [Dan Collins]”

  1. Timstigator says:

    This dealership closing scheme just makes my blood boil. And then to know the MSM won’t pick this story up makes my boiling blood turn to vapor.

  2. Timstigator says:

    Hello, Mainstream Media!

    Hellooooo!!

    *crickets*

  3. HarryB says:

    An analysis of the 75% of the dealerships remaining open needs to be done before drawing any conclusions. If their contributions to Obama were significantly greater than those of the 25% that are being closed, on a relative basis, political hanky panky was probably involved. Without that analysis however, one can’t dismiss the possibility that Chrysler car dealers primarily contributed to Republicans, as a group. In which case, the same outcome would result regardless of the 25% chosen for closing.

  4. Dan Collins says:

    Where your analysis goes perfectly wrong, HarryB, is where you say “the same.” That is bullshit.

  5. B Moe says:

    Even on the remote chance Obama is doing something that isn’t purely politically motivated, under what business model does slashing advertising and closing dealerships lead to more car sales?

  6. N. O'Brain says:

    “…under what business model does slashing advertising and closing dealerships lead to more car sales?”

    The fascist one.

  7. Carin says:

    I will never-ever buy another Chrysler product. Never ever. And, my last minivan (before my current one) was a good car. served me well.

    My next car is a Ford. I like those “Flex” things.

  8. Carin says:

    HELLO UAW. FUCK YOU. Fuck you Ron Gettlefinger with a swordfish.

  9. B Moe says:

    I used to have one of these.

    Basically a piece of shit but very high funk factor.

  10. Carin says:

    Ha. Those are badass. I used to consider them the ugliest car out there, but when I see ’em now I think they’re pretty darn cool.

  11. Mr. Pink says:

    BMoe I own a freakin Convertable Sebring. I never see anyone under 60 driving one of those things.

  12. B Moe says:

    I have a convertible LeBaron, too, but it got trashed by vandals over at my shop, so it is headed for the scrapyard.

  13. HarryB says:

    Dan,

    By “the same” I mean that the 25% chosen would mostly be Republican contributors, most likely, not that the same quantitative result would be realized.

    Please don’t get me wrong here. If the dealers being forced to close were selected because of their political contributions, that’s wrong and needs to be called out. But I don’t think that can be concluded by looking only at the dealers being forced to shut their doors. And if the other side later comes out with data showing that car dealers are predominantly Republicans, it will impact credibility.

  14. LTC John says:

    B Moe, My folks spent many years buying Chryslers to help support the economy of the area they lived in – the next town over had a Chrysler plant. The LeBaron was the car that ended that deal/solidarity… heh. I fully understood the term “lemon” after that hunk of #$%&.

  15. jon says:

    If a bomb goes off at a daycare, most of the dead will be children. If a plane going from India to China crashes, most of the dead will be Asian. If a tour bus crashes in Iowa, most of the injured will be overweight. And if a bunch of car dealerships close, most of them just might have Republican donation histories.

    Conspiracy? Probably not. I can’t think of any major car dealers that didn’t contribute to Republican causes, Republican candidates, or just plain acted like Republicans, and I live in a Democratic area of a Republican state (Arizona). Do I wonder if something fishy may be up in these dealership closings? Hell, yes. Do I think anything has been proven yet? Hell, no. Just looking at the maps of closed dealerships, I see that there are probably too many in the Rust Belt and the Retirement Belt. Individual dealerships will howl about their sales numbers and history and such, but I don’t think anyone looking to buy a car will be unable to find one.

    Also, I don’t know where to begin to criticize when I read analysis–even statistical analysis that involves correct math–that starts as this one does: “All other factors being equal, what are the odds that RLJ’s dealerships would remain open while all other area dealerships would be shuttered? The approximate odds of such an occurrence can be calculated. 789 of the Chrysler’s dealerships are closing, which represents 25% of the total (according to MSNBC).” Nice job there. Take an assumption, add some math, and prove nothing. But look! it has math. But does it prove anything? No. All other factors being equal? Where did that come from? I can take statistics and prove how unlikely it was that the Twins would have won the 1989 World Series, but that would prove nothing. Using statistics with limited information is like using a yardstick to measure a unicorn.

  16. Mr. Pink says:

    The problem is you have politicians making the decisions on which stores close and which stay open, ie central planning of the economy. What ever happened to the “seriousness of the allegation” or “the appearance of impropriety”? I guess those went outa style circa January 2009 like so many other things.

  17. Mr. Pink says:

    16
    WTF. Are you really trying to say “Well I don’t know any dealership that isn’t owned by a Republican.”? Seriously?

  18. Dan Collins says:

    Well, then, the answer is once again to get all the information and then to conduct a stringent statistical analysis. I’m hoping that this kind of reporting will lead to that.

    The disturbing thing, for me, is that Chrysler seems unable or unwilling to disclose the formula that they used to determine which dealerships would close, and which would stay open with expanded territories–assuming there was a formula at all.

  19. Carin says:

    Meya would just like you to look away, because she’s SURE nothing untoward is occurring.

  20. That’s the problem: we still don’t know who’s really pulling the strings on this–whether it’s coming more from the White House, or more from Chrysler. If it’s coming from the White House, we have a right to know what the reasoning was . . . hell, given the fact that we now own a company most of us didn’t want to buy, we deserve to know irrespective of who is “really” making the decisions, WHAT THE CRITERIA WERE. Even if the formulae were different in different regions, and different types of markets, we deserve to know THAT.

    I don’t want to be told that this is over our sweet little heads; I want to know WHY five-star dealerships are being closed down. I want to know how the “Honda-Toyota” theory of automobile distribution captured the imagination of American car companies, and what data are brought in to support it, and who had bought into it before the Obamanoids took the company over.

    And I want to know why a company that I was forced to buy is shutting down revenue streams.

  21. meya says:

    “The disturbing thing, for me, is that Chrysler seems unable or unwilling to disclose the formula that they used to determine which dealerships would close, and which would stay open with expanded territories–assuming there was a formula at all.”

    Yeah, disturbing that Chrysler doesn’t want to disclose statistics on the competitive strength of its remaining dealerships and the business process they used to assess them. What possible reason could they have for that?

  22. SBP says:

    That’s if you were picking random dealerships to close completely independent of the others. Nice mathyness there.

    Why don’t you tell us how they were picked, SFAG?

    Nice lying crapweaselyness htere.

  23. SBP says:

    Yeah, disturbing that Chrysler doesn’t want to disclose statistics

    “Chrysler” doesn’t exist any more, SFAG.

    It’s a branch of the Obama administraton.

  24. Dan Collins says:

    Considering that we are all presumably shareholders in Chrysler, meya, they have a fiduciary duty to disclose the information.

  25. Dan Collins says:

    But, of course, in meya-world that information should be much more closely guarded than redacted “torture” memos.

  26. meya says:

    “Why don’t you tell us how they were picked, SFAG?”

    I’d certainly hope it wasn’t random. That sounds like a rather dumb way to do it.

    ““Chrysler” doesn’t exist any more, SFAG.”

    Thanks for lying to me, Dan Collins.

    “Considering that we are all presumably shareholders in Chrysler, meya, they have a fiduciary duty to disclose the information.”

    Last I checked the treasury was a shareholder, not ‘we all.’

  27. Dan Collins says:

    Meya, your evasions are torporous. Whose money is that?

  28. jon says:

    Mr. Pink, yes I am. I honestly don’t know of any major car dealerships in my area that don’t contribute to Republicans. Is that true everywhere? I don’t know. All I know is that big car dealerships are major players in the political dealings of my area, they contribute to Republicans and Republican business groups such as pro-growth construction groups, and they also tend to be Right Wing dorks when I hear them talk about issues other than trucks and cars on various local radio and television. It should also be noted that none of the dealerships in my area are closing. So yeah, I am using “anecdata” to attack someone else’s assumption when I say nothing has been proven yet. The math used is accurate, but the question is whether or not the formula is correct. I can’t see that it is, nor can I say it would prove the existence of a conspiracy if the other 75% is not analyzed as well.

  29. SBP says:

    Weak, SFAG, even for you.

    I ask again: do you REALLY believe that anyone here is stupid enough to be taken in by your dullwitted dissembling?

    What do you get out of it? Is someone paying you to do it, or is it simply some kind of mental illness?

  30. Dan Collins says:

    Jon, is it fair to say that they need to release the formula that they (presumably) used to make the determinations?

  31. SBP says:

    I see that ObamaChrysler is after another half billion of our money to build “electric cars”.

    No details on who these “partners” are, naturally.

  32. Dan Collins says:

    Meya, you ARE the weakest link.

  33. SBP says:

    Tell it to all the pension funds that had their money stolen, SFAG.

  34. SBP says:

    Meya, you ARE the weakest link.

    Fixed it for you, Dan.

  35. SBP says:

    Meya, you ARE the weakest link mind.

    Fixed it for me.

  36. SBP says:

    Do you know how ridiculous it was the first time I typed those numbers into a binomial coefficient calculator?

    What’s ridiculous is the idea that you understand what you were doing on a level above that of a savage offering a sacrifice to an idol.

    Give it a rest, SFAG. Nobody here is buying what you’re selling.

  37. Dan Collins says:

    “The GAO was established as the General Accounting Office by the Budget and Accounting Act of 1921 (Pub.L. 67-13, 42 Stat. 20, June 10, 1921). This Act required the head of GAO to “investigate, at the seat of government or elsewhere, all matters relating to the receipt, disbursement, and application of public funds, and shall make to the President…and to Congress…reports (and) recommendations looking to greater economy or efficiency in public expenditures” (Sec. 312(a), 42 Stat. 25). According to GAO’s current mission statement, the agency exists to support the Congress in meeting its constitutional responsibilities and to help improve the performance and ensure the accountability of the federal government for the benefit of the American people. The name was changed in 2004 to better reflect the mission of the office.[1] While most other countries have government entities similar to the GAO, their focus is primarily on conducting financial audits. The GAO is unique in that its auditors conduct not only financial audits, but also engage in a wide assortment of performance audits.”

  38. Dan Collins says:

    Despite your evasions, meya, ALL AMERICAN TAXPAYERS are ENTITLED to this information, even under the Transparency Regime of Obama.

  39. Pablo says:

    Even on the remote chance Obama is doing something that isn’t purely politically motivated, under what business model does slashing advertising and closing dealerships lead to more car sales?

    B Moe, you’re laboring under the dated illusion that Chrysler exists to build and sell cars. No, the new Chrysler has a diversified business model: healthcare, pensions, political activism. On the off chance that another car gets made with a Chrysler nameplate, that will have been outsourced to Fiat.

    Think about it. “Fiat trucks are Ram Tough” Is anybody gonna buy that? No. Building it would be a waste of sorely needed resources.

  40. meya says:

    And if you want to get the GAO to act, try to find a member of congress to send them a request. Also have them ask the Congressional Research Service to look into it. You may not get the info, but you’ll at least have someone with some independence looking into it.

  41. alppuccino says:

    Any car whose name starts with “Le” should be avoided.

  42. Pablo says:

    But if you’re not currently a Chrysler shareholder, I’m sure you can just buy a stock for real cheap anytime now and then claim your fiduciary right to this information.

    No, you can’t. I’m glad you’re sure, though. That must be comforting. How does wrong feel?

  43. jon says:

    Dan, I have no problem with the idea of them releasing the formula they used. My guess is that it was less math-oriented and much more a result of a bunch of people looking at a map of all the dealerships and making geographic arbitrary decisions rather than political or even economic arbitrary decisions. But yes, I really have no problem with anyone asking why.

    I do have serious problems with those who analyze things they pretty much admit to knowing little or nothing about. It’s quite possible there’s a stronger link to the letter E than politics, but until the formula is known, I welcome various people wasting their time trying to prove things with limited data. It allows for some fun popcorn time as I watch wild accusations fly, see competing conspiracies fly this way and that, and generally make wingnut bloggers look smarter and smarter as the days pass. And the end result: nothing is solidified other than the conspiracy-theorists’ sense of self-importance. I hope the Chrysler Truthers live long and prosper in their own fun way. I guess trying to melt a girder in a backyard fireplace still looks crazier than a bunch of statisticians with incomplete formulas trying to prove a conspiracy, but please keep trying.

  44. alppuccino says:

    LeBiden

  45. Buckeye says:

    Three possibilities: 1). The Chrysler execs and/or government authorities made decisions consistent with a statistical model that sorted between “winners” and losers” based on strictly economic assumptions (not political); 2). The powers-that-be used the model for the most part, but made some or many exceptions based on political expediency (for example, special friends like RLJ would be assisted, while special non-friends like Jim Anderer will singled out for elimination); 3). No model was ever used and the decisions were based on other arbitrary reasons.

    If scenario #1 happened, then Chrysler and/or the car czar have nothing to hide. It would benefit them to transparently demonstrate that their decisions were based on a fair, business-maximizing model. And, really, what would it cost them to release it, assuming no shenanigans? On the other hand, if either scenario #2 or #3 happened, they do have something to hide.

    It’s a good thing we can count on the watchdog media to dig up the facts.

    I do agree, though, that it is important to compare the political contribution patterns of the saved dealerships versus those selected for elimination. But another good place to start would be to evaluate RLJ dealerships to compare their sales volumes, customer service scores, and local market share versus their competitors (to be eliminated) within the same market areas. If even one of the RLJ dealerships looks like it should have been selected as a “loser” by an impartial statistical model, then further investigation would be warranted.

  46. SBP says:

    No, the new Chrysler has a diversified business model: healthcare, pensions, political activism.

    Don’t forget continuing to fork out $100K/year in “wages” and benefits to auto “workers” who have been sitting on their asses in the “job bank” every day for 20 or 30 years.

    We mustn’t forget their irreplaceable contribution to the American Dream. You think it’s easy to get up in the morning, go to the plant, clock in, play cards for eight hours, and then go home? I beg to differ. And all that for a measly hundred grand per!

    If a few old ladies have to lose their pensions, well, we must all be prepared to make some sacrifices in these difficult times.

  47. alppuccino says:

    LePrompter

  48. Carin says:

    jon, do you mind sharing where you live and who these big time Republican dealers/donors are?

  49. Dan Collins says:

    Sorry, jon, that doesn’t cut it. Meya will be happy to know that I just put in a call to the GAO. Tell me about arbitrary decisions, jon. Would it be comforting to know that the company’s restructuring involved arbitrary decisions?

  50. Carin says:

    I do have serious problems with those who analyze things they pretty much admit to knowing little or nothing about.

    Yes, ’cause that math stuff is SO HARD. It’s best blogger types stuck to simpler stuff. Like talking about Michelle’s GREAT GUNS

  51. Carin says:

    Imagine that Bush had taken away family businesses. Informed them that a company they built or bought and depended on was no longer “theirs” and they would get absolutely no compensation for it. Imagine the howling. But here, now, under Obama – they don’t even like us asking question about it.

  52. Bob Reed says:

    It’s just the Chicago way, that’s all…

    Now if the properly filled brown envelopes had been in the mail…

  53. Mr. Pink says:

    Carin but jon told me all those people are “Big time Republicans” so fuck em.

  54. jon says:

    Carin, I live in Tucson. No dealerships are closing in my area. But the big dealerships are owned by Holmes Tuttle, Jim Click, and Bob Beaudry. All donate mostly to Republicans and Republican-leaning groups and causes. Though I think his politics suck, I buy from Jim Click because he sells Fords, and because he hires lots of disabled people to work at his dealerships. (I like Fords, and I think he’s nice to do that other stuff.)

    As for arbitrary decisions, I have a hunch. Just as the creator of that formula had a hunch that all the items were equally-weighed in the formula used by the decision-makers. I’ll be only too glad to be proven wrong, but until some facts come clear I can’t say my hunch is any better or worse than any other. And just because the math was correct doesn’t prove an incorrect formula gets the correct result, Carin. Get the right numbers in the right columns weighed the right way, and then the math may be hard or simple but it’s much more likely to lead to something that might actually show something. Until then, the statistical “analysis” proves that someone is capable of statistics, but the analysis part needs better data.

    But yes, let’s get the information out there. I’d like to know how the decisions were made. I’m still betting on geographical arbitrariness, but open to being proven wrong. I’m also open to being proven right. Hell, I just want to know.

  55. N. O'Brain says:

    “That’s if you were picking random dealerships to close completely independent of the others.”

    That, ladies and gentlemen, is the sound of a point whizzing right over someones head.

  56. The Other Ed says:

    Guess what guys, 90% of Auto Dealers give more money to the GOP than to the Dems, so odds are 90% of the dealerships being closed would also have given more to the GOP. If it was the other way around and most dealerships being closed WEREN’T Republican, then you would have something to be suspicious about.

    Here’s Nate Silver, a statistician from the Baseball Prospectus, who also blogs on politics, has pulled the numbers:

    http://www.fivethirtyeight.com/2009/05/news-flash-car-dealers-are-republicans.html

  57. Mr. Pink says:

    I have owned 6 different cars and do not know the political leanings of anyone that sold them to me. I guess I am strange that way.

  58. Mr. Pink says:

    I am sure glad this genius is in charge of car companies.
    http://www.nytimes.com/2009/05/28/world/europe/28queen.html?_r=3&ref=world

  59. Mr. Pink says:

    Sorry for the multiple posts in a row but here is another article on this Chrysler stuff.
    http://www.washingtonexaminer.com/opinion/blogs/beltway-confidential/Furor-grows-over-partisan-car-dealer-closings-46261447.html

  60. Rob Crawford says:

    My guess is that it was less math-oriented and much more a result of a bunch of people looking at a map of all the dealerships and making geographic arbitrary decisions rather than political or even economic arbitrary decisions.

    The people making the decisions are Obama administration croneys. The odds the decisions are not politically motivated are nil.

    How do you explain all of the dealerships owned by a Democrat big-wig remaining open, while all of their local competitors being shut down?

    But yes, let’s get the information out there. I’d like to know how the decisions were made. I’m still betting on geographical arbitrariness, but open to being proven wrong. I’m also open to being proven right. Hell, I just want to know.

    No, you don’t. If you did, the RLJ-McLarty-Landers case would be a screaming siren for you. Instead, you’re content to be lulled to sleep.

    Oh, and you’re dodging a critical point: where the fuck did the government get the authority to interfere in the whole fucking mess?!

  61. Carin says:

    . Car dealerships need quite a bit of space and will tend to be located in suburban or rural areas

    He’s making some rather broad generalities there. Ed’s link.

    And the stats he used was for “dealers” – not Chrysler dealers. It would appear that Nate Silver is being just about as accurate as those he is criticizing.

  62. Carin says:

    I’d like a clear cut answer regarding who choses the dealerships to be closed. Who. Was. It?

  63. Dan Collins says:

    The Other Ed: Okay, now factor in the relative amounts of money given to Dems and Repubs by the UAW, and consider whose money’s being used to bail out their pension funds, and whose getting screwed, and then come tell me that there’s no reason to look for anything.

  64. Dan Collins says:

    And let’s go ahead and accept your 90% figure. Out of almost 800 dealerships, only one appears to have given money to Obama. Now, calculate the odds.

  65. JD says:

    Using an arbitrary geographical selection sounds like a fine way for the automobile task force to restructure. Random decisions while creating a restructuring plan seem like such a great idea.

    Meya is a lying pi-ignert douchenozzle.

    jon – Don’t you think that it might have been a good idea to populate the automobile task force with people that had actual experience in the auto industry? That it was filled with political appointments does not give confidence to the decisions that come crashing down from them.

    Not only should we look at the other 75 percent, we should look even closer at who gets the licenses from the closed 25 percent. If we use jon’s silly thinking, we should see approx $600 in donations to Teh One in that group.

    Fuck off meya. Something about a swordfish comes to mind.

  66. Carin says:

    There is no reason to look for anything, because teh One is perfect in every way. Neither he, nor his administration would ever do anything wrong. To even question, to doubt … well, honestly, it just make us look stupid.

  67. happyfeet says:

    I agree with Carin about the swordfish and also this is creepy, watching people get raped while the Kitty Genovese media does nothing.

  68. happyfeet says:

    did y’all know Fiat had like a spin-off in Spain called Seat? I think it’s part of Volkswagen now. I didn’t cause no one tells me anything.

  69. Carin says:

    The media is a serious industry that cannot sully its reputation with silly issues like this. Now, I wonder what Miss. California is up to …

  70. Rob Crawford says:

    The media is a serious industry that cannot sully its reputation with silly issues like this.

    Certainly. They have too much on their hands, finding some CEO making a bad deal somewhere, to give a rat’s ass about the imposition of fascism.

  71. Buckeye says:

    The left-wing sites are going on defense, pointing out that all persons who contributed to a candidate in the 2008 Presidential Election who listed their occupation as car/auto/automobile/automotive “dealer” tended to be more likely to contribute to Republicans than Democrats, which is really not surprising. The FiveThirtyEight site, via Huffpo, claims that all Republican dealers (not just Chrsyler) outspent all Democrat dealers 88% to 12% in the last election. They are mistakenly calling this the “control group” and then comparing it to the “test group” – i.e. the Chrysler dealers who are being forced to close. They claim that the test group spent 92% on Republicans versus 8% on Democrats, which is not a signficant difference.

    So, there might not be any shenanigans after all; however, the above analysis is far from conclusive. First, the list of ALL dealerships (Chrysler, Ford, GM, Toyota, Honda, etc.) is NOT the control group. The control group is the CHRSYLER dealers who were not selected for elimination. For all we know, Honda and Toyota dealers, which sell non-union manufactured vehicles, contributed to Republicans at a 95% v. 5% rate, while ALL Chrysler dealers only spent 80% v. 20% on Republicans. We need to look at the right control group before we can conclude whether there appears to be a statistically significant difference between the two group’s spending patterns. Second, there could have been numerous examples of politically-motivated decisions and we still might not see statistically significant differences between the control and test groups. For example, maybe the Republican contributors tended to run more competitive dealerships than their Democrat counterparts on a “per capita” basis; in that case, if the elimination list achieved proportional parity between Republicans and Democrats that would actually indicate a bad selection process. Moreover, if the Chrysler execs and/or feds for the most part followed a formula, but made a handful of polically-motivated decisions, their choices probably wouldn’t show up as a general pattern, but they would still be wrong and should be investigated.

    I understand the need for confidentiality. Allow a group of Republican Senators to have access to the selection model and data and then they could decide whether the case warranted further investigation.

  72. Dan Collins says:

    Fair enough, Buckeye. But do you see the need for confidentiality regarding who made the decisions?

  73. Carin says:

    Here is another really interesting post about the Michigan dealers which are closing. No minority-owned dealers are closing (there are two) in the Detroit area.

    Apparently, it was feared that minority owned dealers would lose out because they tend to be smaller and carry less brands. Among all Chrysler dealers, 25% were closed. Among minority owned? Only 20%

    But politics had NOTHING to do with it. It was arbitrary.

  74. Jimmy says:

    The mainstream media won’t listen, they have obama/pelosi/biden ears, and they unable to hear any common sense. They are in power now, so to hell with everyone else.. Welcome to our Democratically run United Soviet States of Obama

  75. Mr. Pink says:

    Well that is good to know Carin. I hope that our newest SC justice will go about helping them out at the expense of others even more. They DESERVE it.

  76. Mr. Pink says:

    Sorry I am extremely bitter right now. My company just lost the contract I am on and I know have to go job hunting. Hooorayyyy now I know how Chrysler dealers feel.

  77. Dan Collins says:

    I’m sorry about that, Mr. Pink. Is your resume online?

  78. The Monster says:

    Thus, the odds that any, randomly selected, single dealership would remain open is roughly 75%. The odds that a single dealership would close is roughly 25%.
    . . .
    In Lee’s Summit, MO, .75 * .25 *.25 *.25 *.25 = .0029296875.

    The above is wrong so many ways I can’t begin to describe how wrong it is, so instead I’ll try to demonstrate it:

    There are 347 NCAA Division I schools, so the odds that North Carolina would win the Men’s Basketball tournament and all the others would lose is 1/347 * (346/347)^346 = 0.106%

    You can’t just multiply probabilities that are dependent upon one another. The probability that the other four dealers near Lee’s Summit would get to keep their dealerships is not 75%, once the decision has been made to revoke this one, just like the probability that Michigan State would NOT win the NCAA tournament was linked to North Carolina’s probability of winning. The same event (championship game) determined both outcomes.

    We really need a lot more information than this to come to any conclusions.

  79. Mr. Pink says:

    Yeah it is online. I should be able to ship it to the company that won the contract and hopefully get a job but who knows in this economy.

  80. happyfeet says:

    I’m sorry too Mr. Pink … my little brother got laid off this week and it was directly cause of laws what the dirty socialists passed.

  81. Dan Collins says:

    So, can I flog a link, or something?

  82. Carin says:

    Pink you have every right to be pissed.

  83. Mr. Pink says:

    What does flog a link mean? Sorry I am illiterate when it comes to internet slang. If that means you want my resume I will email it to you if you know of any government or contracting jobs in D.C. or Northern VA that pay at least 60 grand a year.

  84. Dan Collins says:

    If it’s online, I can link it with a short description of your background and also tweet it on Twitter.

  85. SBP says:

    Sorry to hear that, Mr. Pink, and I hope you find something soon.

  86. Mr. Pink says:

    I should be fine though. I’m going to apply at the company that won the contract. If all else fails I could reup in the Army.

  87. DrSteve says:

    I’m afraid meya has a point here. The framing she(?) applied to the statistical problem is the correct one. It’s like asking the question about the probability of selecting any single element from a continuous probability distribution. Any particular choice is going to seem unlikely.

    Chrysler should maintain a state-level dealer list that would allow us to cross-reference against political contributions in FEC databases, to determine whether there’s even any variance in the underlying data (i.e. heterogeneity among dealers).

    There’s really only one way to decide this issue: Get all the data for all the dealerships, to include the closing decision (indicator) as the dependent variable and the factors Chrysler stated it used (plus political donations) as the explanatory variables in a logistic regression. The question then becomes whether the political donations are still a statistically significant explanatory variable when Chrysler’s claimed decision elements are considered.

    I won’t say the political-philosophical-Constitutional issues are not relevant (I think they’re highly important, and Obama concerns the hell out of me) but they’re not relevant to this particular analysis.

  88. DrSteve says:

    Hey Pink, what industry/agency/sector of government contracting did you work in? I’m in DC and in that business, and as you probably know some agencies are getting literally more money than they know what to do with.

    Dan, any way I can pass an e-mail to Pink through you?

  89. SBP says:

    There’s really only one way to decide this issue: Get all the data for all the dealerships, to include the closing decision (indicator) as the dependent variable

    Or we could simply get them to state the criteria they used to make the decision.

    Statistical wanking is better than nothing, but it’s a poor substitute for fact.

  90. Dan Collins says:

    DrSteve, if you read my comments, you’ll see that I’m asking for the information that would be necessary to conduct a rigorous statistical analysis. As for your last sentence, that remains to be seen, doesn’t it?

    At root, the issue is: what were the criteria and who made the decisions?

  91. Buckeye says:

    Dan, no need for confidentiality regarding who made the decisions. They should be completely transparent regarding the entire executive decisionmaking process. As for whether the selection model and data should be made avaialble, maybe that depends. I haven’t seen the breakdown. Is the proposal to make Chrsyler majority taxpayer owned? If so, everything must be transparent, including all the internal data. If that means that publicly-owned companies have a disadvantage compared to privately owned companies, good! They should be created as temporary, emergency arrangements, or not at all.

    In any case, with the infusion of TARP money, we must demand that the selection model and data be auditied by persons who have no incentive whatsoever to curry the favor of the Obama administration or Chrysler. On the other hand, if there was no sophisticated selection model used – as jon suggests, people sitting in a room scribbling on hard copy maps – that would just be ridiculous, if not as corrupt.

  92. Rob Crawford says:

    Bullshit. The sole criteria is “do we want to be free, or do we want to be slaves”.

    Arguing over the statistical methodology of figuring out if a group of political appointees made decisions based on politics is rearranging the deck chairs while the Titanic is steered into the ice bergs.

  93. Dan Collins says:

    No, Buckeye, the proposal is to make Chrysler principally owned by the UAW’s pension fund at the expense of the American people and of the shareholders, then to have those shares sold to Fiat.

  94. Carin says:

    I’m with Rob.

  95. Dan Collins says:

    Yes, I think. Mr. Pink and you can both email me at vermontaigne-at-gmail-dot-com. Thanks very much.

  96. JD says:

    How is this not a takings by the government?

  97. JD says:

    Especially in the cases where the license is being given to a competitor. Or, is this just a silly question?

  98. DrSteve says:

    DrSteve, if you read my comments, you’ll see that I’m asking for the information that would be necessary to conduct a rigorous statistical analysis.

    No, I agree, I saw that.

  99. steveaz says:

    To be fair to Meya, Dr., you’ll need to supply her rationale for Obama’s secrecy on the matter (‘cuz she’s not doin’ it herself).

    There’re damned lies, simple lies and statistics: if sunshine is the best medicine, then why not part the curtains and let it in?

  100. JD says:

    It is prolly racist to even look into this.

  101. DrSteve says:

    you’ll need to supply her rationale for Obama’s secrecy on the matter (’cuz she’s not doin’ it herself)

    Well, I get to say no to FOIA requests from competitors on my government-negotiated indirect rates all the time, even when the contracts I’ve bid them on are a matter of public record and they’re audited by public agencies.

    I agree that there is a legal basis for Chrysler refusing to turn that information over, although in contrast to meya I am not happy about it.

    You could probably do a nonparametric test of independence (donation yes/no, closed yes/no) without any of the covariates we talked about, and based on publicly-available information. It’s nothing I’d run to the AER to publish, but it’s a start.

  102. JD says:

    It is beyond coincidence that none zero zip nada of the big Dem supporters that owned dealerships were cut, and in fact, got increased franchises at other people’s expense as a result of this.

  103. DrSteve says:

    It is beyond coincidence

    Possibly; the question for statistics is how far beyond.

  104. Rob Crawford says:

    Possibly; the question for statistics is how far beyond.

    WTF does statistics have to do with it when the decisions are being made by appointed political hacks?

  105. DrSteve says:

    If someone can compile an exhaustive list of all Chrysler dealerships as of the beginning of 2009 (or some other fixed recent date), and go through the FEC filings so we can code the contributions data a few different ways (gave to dems, gave to rs, gave to obama, gave to obama opponents incl. primary opponents) I am happy to crunch, although at that point I think anyone with Excel could do it. I’m thinking Chi-square, and I think we probably have the right number of observations (not so many we overpower the null).

  106. JD says:

    DrSteve – Though I admire your level-headedness, I am not willing to go through mental gymnastics to see that which is readily apparent. None zero zip nada sifr of the big Dem donors were effected by the cuts. Actually, they were effected, just in a good way. Their competition was cut, and they were given increased licenses for their franchises. Couple that with the fact that Gattner is a political operative, with no industry experience, and I simply fail to see what the point of a rigorous statistical analysis will show us that simple common sense is not already screaming at us.

  107. DrSteve says:

    WTF does statistics have to do with it when the decisions are being made by appointed political hacks?

    Let me suggest how this can be instructive. A few years ago I was auditing a couple of federally supported entities. We did an analysis of their errors in complying with regs. One of the analyses we performed was to determine what percentage of errors benefited the entity, and what percentage went against it — and whether this was significantly different from 50 percent.

    Statistics allowed us to say exactly how unlikely was the entity’s explanation that those errors were perfectly random.

  108. JD says:

    Though maybe it will show that the Chrysler dealers, 76-24 split in the last election cycle donations, have been giving more to Dems since Nov, and will clearly be shoveling money to Teh One after this fiasco.

  109. McGehee says:

    I used to have one of these.

    That got me to thinking about the Chrysler-made cars I’ve driven or that my family has had. All were made in the 1960s.

    The one I remember with most fondness was a ’65 Dodge A100 minivan. It’s also the only one of the three that was bought from a dealer rather than through a want-ad. And it’s the only one I never actually got to drive because I wasn’t old enough.

    I remember the summer trip halfway across the country when I was eight, where it took us four days to get across Wyoming because the radiator was crapping out. Good times.

  110. JD says:

    There is surely a math person here that could answer a question like this. In a jar there are 3000 marbles, split at 75% black and 25% white (2250 black and 750 white). What are the odds that if someone began selecting marbles, and picked out 750, that not a one of them would be white? Or, that only one of them would be white?

  111. JD says:

    My prior question was wrong.

  112. JD says:

    There is surely a math person here that could answer a question like this. In a jar there are 3000 marbles, split at 75% black and 25% white (2250 black and 750 white). What are the odds that if someone began selecting marbles, and picked out 750, that not a one of them would be white? Or, that only one of them would be white?

    Never mind. That is exactly what I meant to ask.

  113. Mr. Pink says:

    After looking at the Dr.’s resume and comparing it to my tiny shrivled one I will recuse myself from this conversation if it goes anywhere near measuring statistics.

  114. McGehee says:

    118. Comment by Mr. Pink on 5/28 @ 11:46 am

    Yeah, I was reduced to reminiscing about a minivan.

  115. Rob Crawford says:

    Statistics allowed us to say exactly how unlikely was the entity’s explanation that those errors were perfectly random.

    IN THIS CASE THE ENTITY SHOULDN’T BE MAKING THESE DECISIONS IN THE FIRST PLACE!!!!!!!!!!!

    Goddamn, talk about missing the forest for the fucking trees.

  116. JD says:

    I know this for a fact. It is beyond the idea of random chance that none of the Dem donor fell into the “cut” group, and in fact, all fell into the safe, or even expanded group. I am just a simple minded hillbilly rube, but I cannot imagine a scenario where the politics did not play into this. The numbers are too big.

  117. Rob Crawford says:

    JD, politicians are making the decisions. There’s no way they’re not making decisions based on politics. The back-and-forth on statistics is a red herring.

    It’s like complaining that, pre-Civil War, the majority of slaves were of African descent. So what? The problem isn’t who is enslaved, it’s that there are slaves at all.

  118. jon says:

    As for “How is this not a takings by government?” I’d answer: because they didn’t take anything. Unless they give the dealership to someone else or keep it, it’s not a taking. Closing things isn’t a taking. It’s a business decision. If you don’t want government making business decisions, don’t go bankrupt.

  119. Mr. Pink says:

    Jon Lithia motors has gained dealerships. The did get something out of the deal, but hey “We all have to have some skin in the game.”

  120. DrSteve says:

    JD, I calculate the answers to your questions as follows:

    > phyper(0,750,2250,750) # This is the likelihood of drawing no white marbles in a “handful” of 750 pulled from an urn with 2250 black marbles and 750 white ones.
    [1] 2.219479e-111

    > phyper(1,750,2250,750) # This is the likelihood of drawing one white marble in a “handful” of 750 pulled from an urn with 2250 black marbles and 750 white ones.
    [1] 8.339695e-109

    Did I read your statement of the problem right?

    We use the hypergeometric since we’re sampling without replacement.

  121. JD says:

    Jon – That was aggressively dishonest, even for you. The dealership licenses had value, large values in some instances. And in many, if not all, cases, these licenses are now being reallocated. See the link Mr Pink provided above. How many of these dealers were bankrupt?

  122. Dan Collins says:

    I don’t think that your perspective and JD’s are irreconcilable, Rob. But I do agree with JD that prima facie the prospect that–all other things being equal–the closings have shaken out as they appear to have done is likely infinitesimal. So, there’s a presumptive bias here.

    What the nature of the bias is, is a different matter. Without Chrysler coming clean about its criteria, we are left to speculate. It seems to me that it ought to be a simple matter for them to release the relevant data to statisticians, who can explain their methodology and come to a conclusion. It’s only at that point, really, that we can expect most people to understand the implications.

  123. happyfeet says:

    I don’t see what this has to do with marbles. It’s about freedom.

  124. JD says:

    Thank you, DrSteve. I am not a math guy. Can you translate that idea into 1 chance in 8 for me? Would the answers be different if instead of a handful, they were selected 1 at a time?

  125. Mr. Pink says:

    What freedom, he won. He gets to close dealerships, he gets to fire CEO’s, he gets to spend trillions of taxpayer monies that have yet to be collected, he gets to tell us not to use as much electricity. We do not really have a say in the matter anymore I’m afraid. Hell good luck even getting the most transparent administration evah to release the criteria they used to close these dealerships and keep others open.

  126. DrSteve says:

    You’re looking at 2.2 in 10 raised to the 111th power (Sestrigintillion?), and 8.3 in 10 raised to the 109th power (10 Quinquatrigintillion?).

    Which is funny, since the Sestrigintillion was going to be the replacement for the Caliber. Sigh…

    No difference in a handful versus one at a time, except that in the case of a handful you know you’re not talking about replacing. One at a time could either be with or without replacement. Sampling with replacement requires a different distribution (since each draw doesn’t affect the probability of subsequent draws).

  127. JD says:

    So you are saying that there is a chance … ;-)

  128. Bob Reed says:

    JD,

    Fractions raised to an exponential power result in decreasing orders of magnitude; just do 1/10 squared in your calculator…

    I believe the values that Dr Steve has arrived at translate into layspeak as a snowballs chance in hell…

    Am I right Dr Steve, or do I need a refresher course..?

    Or, were those values of the “one in [insert large number here] variey?

    Best Wishes

  129. Bob Reed says:

    Which is still slim…

  130. Rob Crawford says:

    Unless they give the dealership to someone else…

    If you were paying attention, you’d know this is happening.

  131. Makewi says:

    I’m still trying to wrap my mind around the fact that decisions about what Chrysler dealerships are being closed are being made by an appointed “car czar”, an employee of the executive branch of the United States government.

    Since I can’t even get that far, everything else is just a big pile of fail.

  132. Rob Crawford says:

    What the nature of the bias is, is a different matter. Without Chrysler coming clean about its criteria, we are left to speculate.

    According to some reports, the decisions on who gets their business shut down is being done by the Obama administration, not Chrysler. It’s only been since the question of why certain dealerships are being closed that the word’s gone out that it’s Chrysler making the decisions.

  133. JD says:

    Makewi – Does it make you feel better or worse about the decision knowing that Gattner has no auto industry experience, and that he is married to a DNC fundraiser?

    Did none of you cath the Dumb and Dummer quote?!

  134. […] suggest that Chrysler auto dealers donated 76% Republican and 24% Democratic. So, rough estimating, JD puts the question: Comment by JD on 5/28 @ 11:45 am # |Edit […]

  135. Rob Crawford says:

    Did none of you cath the Dumb and Dummer quote?!

    I’ve avoided watching any of the Obama campaign promo videos.

  136. It seems to me that there are only two criteria to consider. The dealer’s sales and the size of that dealer’s market. In other words, a dealer in the top or bottom 2% of his rural market vs. a dealer in the top or bottom 2% in a large urban market. For instance, advertising costs in a large urban market are astronomical compared to those in a small rural area. Same with wages paid. Since I’ve worked in advertising in both types of areas, I can say for a fact that one full page co-op ad in a large urban paper can be thousands more than in a small town paper. Both dealers may sell the exact same number of cars, but the profits will be very different.

    However, once the formula adjusts for these variants, then it should be an easy decision to lop off those in the bottom 2% or 25% or whatever. But, you can’t compare dealer to dealer without knowing some of the other demographics.

    If you find that a dealer in the top 2% is closed while a dealer in that exact same market that is in a lower bracket remains open, then it is time to look for hanky panky and political considerations.

    As an aside, the last time I bought a new car was in 2005. I went first to our “Mile of Cars,” a mile long strip of wall-to-wall dealerships and made the best deal I could get. At the last minute, I decided to take a day to think about it and I happened to have to make a trip to a desert town about 75 miles away. While I was there, I saw the same vehicle on display at their local dealership and stopped in. The deal I was offered there was nearly $3000 less for the very same vehicle. I bought my car at that dealership on the spot. They even delivered it to my house the next day and with a full tank of gas. This impressed me because a few years earlier, I bought a new car that ran out of gas two blocks after I drove it off the lot, leaving me stranded in rush hour traffic.

  137. Dan Collins says:

    Oh, but Rob, that CAN’T BE. The TARP people told me that it was an internal Chrysler decision.

  138. I posted this video from 1950 a few days ago after seeing the link on Twitter. Take a look. It is an eye opening cartoon.

    http://www.youtube.com/watch?v=dXgUGtwmyq8&feature=player_embedded

  139. Makewi says:

    JD – Neither of those facts makes me feel any better about this situation. Nor does the fact that the UAW was privileged over investors contrary to normal practice in similar situations.

    At the end of the day these companies are still not going to be making products that Americans want to buy. So what exactly is the point of all this?

  140. JD says:

    Amen, Makewi. The whole point appears to be to pay off the UAW, screw the institutional investors, and for Gattner to get his hands on a couple automakers. We have already given them billions, and both are still worse off today than prior to the government getting involved. And it is only going to get worse. Pot odds, and all.

  141. Rob Crawford says:

    screw the institutional investors

    ITYM “screw the rule of law”.

    On some level, the Obamatons have to be stunned with how easily they’ve been able to rule by decree. I wonder if the “you haven’t seen anything yet” remark reflects them realizing it.

  142. Rusty says:

    Doc, with all due respect this is a political decision made by poliiticians. What are the odds there’s going to be a political result?

  143. GaryC says:

    JD:

    Thank you, DrSteve. I am not a math guy. Can you translate that idea into 1 chance in 8 for me? Would the answers be different if instead of a handful, they were selected 1 at a time?

    The typical estimate is that there are roughly 10^80 protons in the observable universe, and the universe is about 4×10^17 seconds old. So if one random proton changed into a pink unicorn for one second, then the chance that you would be looking at it is about 1 in 10^99. This dealer thing is less likely.

Comments are closed.