Cash for Clunkers, Indeed

It wasn’t so long ago that everyone just loved Cash for Clunkers. People thought I was a total crank when I scoffed. Even some of my fellow ideologues wavered.

On my now-defunct blog, I wrote,

Whenever I mention that paying people to destroy vehicles does not create wealth, I am inevitably — always, always, always — presented with one of the positive (but decidedly local) effects of the program, as if it negated the simple destruction of goods:

  • American auto companies will be kept alive. Yes, but we could keep nearly any industry “alive” by paying people to continually destroy its products. This only proves that the industry probably shouldn’t be kept alive anyway. By the time it needs this kind of help, it should be allowed to fail.
  • The environmental effects will be positive. Yes, but they’ll be tiny when we consider the effects of manufacturing and running the replacement car — and you’re still paying people to destroy goods.
  • A lot of those cars were more or less junk anyway. Yes, but you’re making them 100% junk — the program’s own terms demand that they be rendered completely inoperative. How’s that an improvement?
  • …[But] countries can’t grow wealthy by destroying goods in exchange for pieces of paper. If they could, then there would be absolutely no reason to stop at cars…

No, the appropriate course would be to generalize, and to destroy all goods in exchange for government scrip. Then we could play Monopoly, I guess, for what all good the money would do. But we’d have to scrape a board in the dirt to do it.

That’s because money isn’t wealth. Money is at best a measure of wealth, which actually consists of goods. Money retains its value as long as there are goods to be traded for it. When the goods disappear, the economy grows poorer, regardless of how the money is shuffled around.

And the payback isn’t long in coming — today’s used car prices are soaring owing to reduced supply. (This link gives even more dramatic numbers, but I’m less sure of them. h/t Radley Balko.)

See how that works? You can’t get something for nothing. Cash for Clunkers turns out to have been a highly inefficient wealth-transfer program, that is, one that destroyed a bunch of wealth along the way. It gave wealth to those already relatively wealthy people who did the government’s bidding (that is, those who could afford to part with a used car and buy a new one). And now it’s taking wealth from those relatively poor people who need a used car today — in the form of higher prices.

Along the way, it destroyed hundreds of thousands of cars — that’s the real wealth these poor people don’t have access to anymore, because the scrapped cars aren’t a part of the economy.

And this is what passes for a successful government program.

Please do be so kind as to share this post.
TwitterFacebookRedditEmailPrintFriendlyMore options

118 thoughts on “Cash for Clunkers, Indeed

  1. Though I agree on the idiocy of the government program devoted to destroying functional capital… I am unconvinced that the bulk of the price hike can really be blamed on C4C instead of a weaker economy. If C4C were to blame, there would be a greater price hike among older vehicles (more likely to be targeted by C4C) than among the 3-year-old set. Instead, the opposite is the case. That leads me to believe that the primary driver (no pun intended) of all of this is consumer thriftiness.

    Report

    • @Trumwill, I’m with Trumwill, also the article notes that newer cars have come down in price a lot relative to used cars.

      That said I was never a fan of c4c. But it should be considered a successful government program by at least one measure; it ended.

      Report

      • @Simon K, when the economy is bad, there’s a premium for cheap products. If Hormel can’t make enough Spam to keep up with demand, prices get raised. The same goes for used cars.

        Oh, and what Aaron said about decreased supply via people that are deciding not to tri-annually upgrade their cars or elsewise hold on to their cars than they historically have.

        Report

  2. Without dismissing your larger point, the article focuses on late model used cars, not cars with a <$4,500 trade-in value that might have been removed from the market by "Cash for Clunkers". It seems to me that the issue is that people are holding onto their cars instead of trading them in. Other factors are likely significant: people who used to get a new car every two or three years may no longer have the income or credit to do so, the pool of people considering late model used cars over new vehicles has likely increased, fleet sales (a big source of late model used cars) may have been down during the years at issue, auto sales plummeted starting in 2008 meaning that there are fewer cars in the 2 – 3 year old range to begin with….

    Your second link (the one you're less sure of) reminds me that a few years ago new SUV sales and used SUV prices dropped through the floor. A large part of a “30% increase” would be fairly attributed to lower gas prices reviving demand for SUV’s.

    Has a similar increase been observed for the vehicle classes and model years most affected by “Cash for Clunkers” trade-ins?

    Report

    • @Aaron,

      It seems to me that the issue is that people are holding onto their cars instead of trading them in.

      An excellent point. I was looking more at the demand side (people buying used instead of new) but the supply side is likely affected as well, C4C or no. It’s silly to get rid of a car after three years and the current economy is sobering.

      Has a similar increase been observed for the vehicle classes and model years most affected by “Cash for Clunkers” trade-ins?

      That’s what I’d really like to see. There were some strange decisions on which models should be included and which ones shouldn’t. Even among classes, full-size sedans from one company would be included but not another.

      On the other hand, it could be a wash if the price-availability differential simply results in people that can’t get their hands on a car in the first category just going ahead and buying something from the second. Still, considering brand/model loyalty, you would expect some differential I would think.

      Report

  3. I wonder if there are any data pertaining to the increase in aftermarket and remanufactured parts having resulted from so many Clunkers being euthanized.

    Report

  4. Good article! I agree that CFC was a terrible program for the reasons stated. Was shopping for a good used SUV a few weeks ago and found that they are in short supply. Dealers in Colorado appear to be fixing up totaled SUVs in order to create inventory.

    Report

  5. Pingback: Balloon Juice » Blog Archive » Cash for Clunkers fail

  6. Pingback: The Oldest Established Permanent Floating Craps Game in New Yorkes « Wild Boar from the Forest

  7. Money is at best a measure of wealth, which actually consists of goods.

    Not exactly.
    Money is more of a commodity, ie liquidity, and retains a value relative to goods.
    For example, if I have a Picasso original on my wall, does that make me a millionaire? Doesn’t help the checking account much. Hope it makes the occasional cat vomit blend in a bit better (depending on the lighting and my state of wakefulness).

    Food for thought:
    One of my favorite lines from the Minutemen (from “Political Song for Michael Jackson to Sing”)–
    Is your life worth a painting?

    The DOT values human life at $200k when figuring the total amount of damages.
    So how many people can be put to death for one Picasso original?

    Current projected returns for my annuity at full retirement is $850k.
    So, if I retire at 30 years, I will be worth the lives of four persons.
    Just wish I could choose which four.

    Report

  8. When I first saw this post heading, I thought it was about GM filing to go public again. Did anyone else hear about that? The Economist was all over it, but I’ve seen zilch on every other news outlet.

    Report

  9. A little while ago I started looking through your archives here to see if you have ever expressed an opinion that surprised my preconception of you.

    Not once.

    Report

  10. Pingback: Stones Cry Out - If they keep silent… » Things Heard: e135v4

  11. Pingback: Thursday Highlights | Pseudo-Polymath

  12. all this talk, and nobody bothers mentioning that C4C’s stated purpose was to get old and inefficient cars (and trucks) off the road and replace them with more fuel-efficient models ? which it did.

    why would you not judge a program by the goals it was intended to accomplish?

    “And the payback isn’t long in coming — today’s used car prices are soaring owing to reduced supply”

    first, you need to prove that C4C is the primary reason that used car prices are going up. right now all you have is correlation. and you really don’t even have that, since used car prices were going up before C4C started.

    Report

    • @cleek, “why would you not judge a program by the goals it was intended to accomplish?”

      Because the collateral damage of a program can quite easily outweigh its benefits, even if it achieves its stated purpose. This program achieved at most a marginal environmental benefit, while doing so via a de facto wealth transfer from the poor and lower middle classes to the upper middle class and rich. Moreover, because many of the purchases under the program (as attested by the massive drop off after the program’s conclusion) were simply people who moved their car buying decisions up a few months to take advantage of the subsidy, the stimulus effect was quite small and unsustainable – sales dropped precipitously after the program ended.

      That’s not to say that the program had no benefits whatsoever, just that those benefits were not large enough to outweigh the unintended – but entirely predictable – costs.

      Report

  13. Pingback: Looking Back At Cash For Clunkers « Tai-Chi Policy

  14. This is incorrect on multiple levels.

    1. C4C caused only 700,000 vehicles to be turned in. The US fleet at any given time is about 62 million registered vehicles and 6.4 unregistered ones that are nonetheless functioning. Additionally worldwide car production is about 40-50 million per year. The numbers are simply not enough to cause a supply shortage in the used car market.

    2. While I haven’t found good figures about the quality of cars turned in, the economics of the situation implies they should have been really junky. Consider two cases, one where a person has a relatively junky used car worth maybe $500 and the other where his car is worth $4400. The $4500 rebate makes a $4K profit for the first person but only a $100 profit for the second. Logic would dicatate that the actual capital destroyed by the program, which I admit is a real cost, is almost certainly very low and was probably not destined to last much longer anyway.

    3. Even more damming to the ‘increased price’ theory is that 84% of the vehnicles turned in were trucks. If anything the prices of used trucks should be shooting up while used cars should remain relatively untouched if C4C had anything but a trivial role to play in the used market.

    3.1 This may explain why used car prices are higher. With the recession AND the fact that people still remember $4 or $5 gas, people have less need for work trucks and more desire for fuel efficient cars. In the new market this is no problem, the auto makers simply shift production from trucks to cars. The used market, though, isn’t able to shift production. When shopping for 5-10 yr old vehicles you are more or less stuck with the production mix of 5-10 yrs ago.

    4. The average replacement vehicle went from 15.8mpg to 24.9mpg. At 13,000 miles driven per year, that’s a savings of about 300 gallons. At a conservative cost of gas at $3 per gallon ($2.50 market plus $0.50 non-market costs) that’s nearly $1,000 in savings per year. Considering the cost of the destroyed capital is almost certainly not near the $4500 mark, but closer to the $500 mark, chances are the cost of the destroyed capital is already paid for.

    Report

  15. I think given the circumstances C4C was pretty good. It provided stimulus to the economy, did boost domestic auto production (thereby limiting our bailout liabilities a bit) and was a gentle correction to the market (if gas was priced at its true environmental cost, the worst of the junkers would probably be taken off the road). I wouldn’t want to make it a permament program though. If it was it would become more liable to gaming with dealers and buyers narrowing the mpg gains. Claims of its failure are quite overstated IMO.

    Report

    • @Boonton, hey, Boonton.

      How much did C4C cost the gummint?
      Divide that by the number of cars sold because of this program… and compare *THAT* cost to the “dividend” one got for getting relatively recently new used cars off of the roads.

      Still worth it?

      I mean, when you compare it to the fact that car sales fell off a cliff the day after C4C ended making it look like everybody just reshuffled the purchases that they were probably going to make anyway around.

      Report

  16. Recently new used cars? A lightly used car typically costs more than $4500 so that would imply a large number of people cashed in cars for $4500 that they could have sold privately for more. Evidence of that?

    In terms of reshuffling, if a new car costs $20K, let’s say, the cash portion represented a 25% savings. Are you telling me that a 25% cost reduction would result in no additional sales? I don’t think so. More importantly the $4500 serviced as stimulus with an environmental edge. Proper for its time, IMO, but only as a one shot temporary program.

    Report

    • @Boonton, sold privately? Or traded in to a dealer privately?

      You wouldn’t believe the folks out there who, when given the option between Craiglist/Putting a sign on the windshield and going to the dealer for a trade-in value simply take “let’s just go to the dealer and get a trade-in value”.

      Additionally: Hey, dude. It’s me. The Jaybird you already know. Stick around. This is a nice community. It would benefit from your (regular! not once in a goddamn while) presence.

      Report

      • @Scott, Let’s think about it a moment. The value of a certain used car is $500. That would seem to include the sum of the value of its parts as well as the value of the car itself….whatever useful driving you can get out of it before it dies.

        Since the rebate maxed out at $4500 we can assume almost all the cars had to be worth less than that. Since the rebate is worth the most with the least valuable car, it makes sense that people cashed in the low value cars before the higher value ones. Long story short, what about the spare parts?

        Also I could be wrong but I believe only the engine block had to be destroyed so not all spare parts are lost.

        Report

    • @Boonton, Okay, just under $3 Billion.

      That’s $3,000,000,000. How many cars were sold under the auspices of this program?

      I mean, if folks who bought/traded in their old cars got $4,500 and it cost $4,501 per car, that’s fricken awesome!

      And if it cost five figures per car, we’re in yet another situation where the government screwed the proverbial pooch.

      What happened, Boonton?

      Report

          • @Jaybird, Kinda misses the point Jaybird. The program had multiple goals, only one of which was to boost car sales. Another goal was to increase the mileage of cars in the US fleet by eliminating low mpg cars and replacing them with high ones. So even if we assume 125K cars were totally new sales how many of the balance of cars sold resulted in higher mpg than would have otherwise been without the program? Likewise the program was also stimulative in order to tap unusued supply during a deep recession. So if a program had 3 major goals why would you evaluate it on only 1?

            Report

            • @Boonton, well, programs not only have goals, they have costs as well.

              Is it not fair to weigh the goals vs. the costs and say “nope, this was, at the end of the day, something that delivered way too little at a price that was way too high”?

              If it would have been cheaper to just cut a check for $4500 and give that to anybody trading in a car that was X years or older in exchange for a new car with MPG of Y or higher… couldn’t we say that that would have been preferable to a program where people traded in SUVs that got 14/15 MPG for SUVs that got 19/20 MPG at a cost of a hell of a lot more?

              One of the things you have to weigh the program against is “would doing nothing at all have done as much damage”… and if the answer is “less”, it doesn’t really matter how many ostensible “goals” were reached, does it?

              Report

            • @Boonton,

              You can’t tout the program for comprising a tiny percentage of the nation’s auto fleet — hey, no average price increases! — while simultaneously touting how big a change it made in average gas mileage. A tiny number is a tiny number.

              And, while I was fully willing to give you the first point, now I realize that you’ll just say anything you can find to defend it. Not very convincing.

              Report

  17. Jaybird.

    “Is it not fair to weigh the goals vs. the costs…”

    Yes but it’s not fair to just look at one goal. Doing so, clearly, inflates the calculation of ‘cost per unit of goal achieved’ in a deceptive manner. To use an analogy, imagine a bar that invests $20,000 in multiple flat screen TV’s with the twin goals of bringing more customers to the bar per night and motivating them to spend more per drink. If 4 new customers appear at the bar you could say this investment cost “$5000 per new customer gained” and declare it a failure, but that wouldn’t be accurate since you’d also have to include your other goals in the calculation. Namely, did the old customers spend more on drinks than they did before or not?

    If it would have been cheaper to just cut a check for $4500 and give that to anybody trading in a car that was X years or older in exchange for a new car with MPG of Y or higher… couldn’t we say that that would have been preferable to a program where people traded in SUVs that got 14/15 MPG for SUVs that got 19/20 MPG at a cost of a hell of a lot more?

    I’m not seeing how that would have been cheaper. At the end of the day you’re cutting a check for $4500 per car for 700,000 cars. If people brought more expensive new cars under the program as opposed to the hypothetical one that doesn’t really alter the cost of the program as the only thing the program pays is the $4500.

    As for making it based on years, I’m not seeing why that would have improved things. As I pointed out the average trade in was significantly better in mileage. The savings per car is running near $1,000 per year or even more depending on what you consider the costs of oil uncaptured by the market price.

    Now in contrast if the program had only one goal, that of boosting new car sales, it would have been a straight forward $4500 per new car, first come first serve when the $3B is up the program is done. since when you’re in a depression just about any form of stimulus works this too would have been ‘costless’ in the sense that it would have offset its cost with economic growth…but the program as it happened had the nice added effect of boosting the US fleet’s mpg which comes in handy when the economy is at full employment.

    One of the things you have to weigh the program against is “would doing nothing at all have done as much damage”… and if the answer is “less”, it doesn’t really matter how many ostensible “goals” were reached, does it?

    It depends on how valuable the goals were compared with the costs of the program. IMO the only cost of the program was the lost capital in destroying some used cars that still had some useful life in them. As I pointed out I suspect the mpg savings alone offsets that loss since the used cars were almost certainly at the low end of the market.

    Now just as a word of caution keep in mind I said I thought the program was the right thing at the right time and the right size. I’m not sure I would have liked it as much if it was $30B or $300B instead of $3B or if it was in place for 5 years or 10 years instead of 1 year. At the scale it was in place I think the program took advantage of a market ineffeciency. Gas is not correctly priced in our economy. In addition to explicit direct and indirect subsidies, there are real costs to using oil that are not captured in the market price of gas. That means that some of those high mpg used cars were on the road because gas is too cheap. If gas was correctly priced those high mpg used cars would have come off the roads faster and made it to the junk yards to get turned into higher mpg cars. So in a way this program would be an example of a gov’t regulation offsetting the harm done by other foolish gov’t programs. In the ideal world I’d simply have gas priced correctly and let stimulus be more ‘pure’ (say in the form of a payroll tax holiday coupled with unemployment benefit boosts).

    Report

    • @Boonton, it’s also not fair to look at just one cost either. We have the cost that is the outlay from the government. We have the cost that is the destroyed wealth (the “clunkers” that would have made better used cars for a number of people out there than what they were already driving… the Datsun Pickups, for example). You have the costs associated with the sales falling off a cliff the moment the incentives ended. (Something we’re also seeing with housing now that the 8 grand credit is gone.) I’m pretty sure that there are additional costs that I’ve overlooked… such as the opportunity costs of “could this exact same money have been used in a more effective manner to help more/different people rather than, say, the folks in the market to purchase a new car?”

      I think it’s quite fair to reach the conclusion that the price was way too high for what we got… even taking the fact that people can now drive 16-25% more than they used to in a week while having to fill up only as much as they used to.

      (The number one thing bought by cash for clunkers? SUVs. The number one thing traded in? SUVs. It was an SUV upgrade program. Was that what was intended?)

      Report

      • @Jaybird, “We have the cost that is the destroyed wealth (the “clunkers” that would have made better used cars for a number of people out there than what they were already driving… the Datsun Pickups, for example).”

        I agree with you here. The cars were destroyed when they could have provided us with a few more years of useful service. But this value I believe was pretty low (let’s say $500 absent any real numbers). At reducing gas costs $1000 per year we’ve already recupped that value.

        You have the costs associated with the sales falling off a cliff the moment the incentives ended.

        Your source said 125,000 additional sales happened that wouldn’t have happened so what ‘cliff’ are you talking about and what costs are you talking about? During a period when there was massive unemployment in the auto industry and auto sales sector what’s the cost in making use of unusued resources? were those auto salesmen performing some vital national service by standing on the unemployment line?

        http://online.wsj.com/mdc/public/page/2_3022-autosales.html

        Note that car sales have been pretty steady. Assuming the blue spike above the 6M mark represents the C4C boomlet, there really isn’t evidence of a post C4C crash in new auto sales. First of all, 700K is not a big deal when car sales are running at 6 million a year (the graph can be confusing, it’s not 6 million a month but an ‘annual rate’ by month with a seasonal adjustment). More importantly, if this was just about demand shifting then we’d see the annual rate of car sales return to the low before the blue spike, it seems to have returned to a level above (we went from an under 6m per year rate to a slightly under 6m per year rate).

        Your cost might have been important if the auto industry was at full employment where the demand for cars would have meant adding temp. workers and training additional workers to fulfill the spike in demand only to let them all go once it disappeared. But at $3B and 10% unemployment the program was simply not large enough to have important effects in that regard.

        I’m pretty sure that there are additional costs that I’ve overlooked… such as the opportunity costs of “could this exact same money have been used in a more effective manner to help more/different people rather than, say, the folks in the market to purchase a new car?”

        Is this kind of like the opportunity cost I’m presented when I consider whether to put $1 against my electric bill or use the $1 to buy the winning lottery ticket? Yes opportunity cost is real but you really need to take it with a grain of salt and there’s a good reason to keep it a seperate consideration than accounting cost. Perhaps there were other places there the gov’t has spent $3B to better effect and I’m sure in the world of hypothetical universes where Jaybird rules as the Ever Wise Philosopher King $3B would be employed in even better ways. That’s an important thing to consider but not to get too hung up on. In other universes, for example, the US won WWII at a much lower cost. That’s important but doesn’t justify calling WWII a military failure for the US.

        I think it’s quite fair to reach the conclusion that the price was way too high for what we got… even taking the fact that people can now drive 16-25% more than they used to in a week while having to fill up only as much as they used to.

        Maybe we do but so what? I figure we incurred a cost of $500-$1500 and are saving about $1K per year, maybe more. We can spend that $1K on anything we please and yes some of that may be doing more driving. Although again this is just speculation on your part. If I got a car with more mpg I might do one or two more road trips in a year but I’m not going to seek a more distant work commute. I’m unlikely to increase my driving by 25%, I don’t enjoy it that much.
        This nifty chart seems to confirm that:
        http://www.nytimes.com/imagepages/2010/05/02/business/02metrics.html

        Not only are we driving fewer miles than in 2007-2008 but 2010 seems to be less than 2009 even though the economy is slightly better. Keep in mind, though, that C4C only impacted 700,000 vehicles or so out of a fleet of over 60 million. I don’t think anyone really has evidence that C4C drivers have dramatically increased their miles driven nor will you be able to see much of a shift in the marco-variables.

        (The number one thing bought by cash for clunkers? SUVs. The number one thing traded in? SUVs. It was an SUV upgrade program. Was that what was intended?)

        I don’t know but the average trade resulted in going from a 15.8 mpg vehicle to a 24.9 mpg vehicle and 84% of the vehicles turned in were trucks which tend to be worse than SUV’s.

        Report

  18. You can’t tout the program for comprising a tiny percentage of the nation’s auto fleet — hey, no average price increases! — while simultaneously touting how big a change it made in average gas mileage. A tiny number is a tiny number.

    Yes I can. The program was a tiny program. If I paint my house I dramatically improve its appearance but in terms of my entire neighborhood or entire state the impact won’t even be noticeable. That doesn’t make my ‘house painting program’ a failure.

    I’d evaluate my program with a scale that’s proper for its scale. If I paint my house I’d expect a dramatic improvement in my house. I’d expect a modest improvement in the overall appearence for the block. A trivial improvement for the town and beyond that I wouldn’t expect much of anything that could be detected so why would I try to measure my program by asking how does all the houses in my state appear? Likewise if you have a program that maybe impacted less than 1% of the US fleet why evaluate it based the average of the entire US fleet?

    That would be a proper metric if we were talking about a program that was $300B….but we aren’t. This reminds me of the bashing Obama took during the election over the Annenberg Challenge that supposedly wasted the Chicago school system (because of the evil influence of Bill Ayers). But when you looked at what was actually done in Chicago it was only a handful of individual schools that implemented some pilot programs….nothing that you could expect to show up by looking at systemwide results data.

    Now would it be nice to improve the entire US fleet in mpg by as much as the 700K cars in C4C? Of course, but as I said I think if you tried to scale the program up by, say, multiplying it by 100 you’d start running into problems which would bring your success metric way down. (But that’s true of house painting too. My house may get a lot better with a fresh coat of paint but would the entire real estate stock of the US improve as much by fresh paint? Probably not)

    Report

      • @Jason Kuznicki, In what way? I pointed out the run up in used car prices cannot be pinned on C4C because too few vehicles were impacted. Jaybird implied that those who got more efficient vehicles in C4C offset their improved mpg with more driving. My stand on that is that there’s no evidence of that. Now fleetwide we aren’t seeing more miles driven, we are seeing less. I assume over the entire fleet we are seeing better mpg averages as new cars slowly replace older ones but I didn’t say C4C shifted the entire fleet in a noticeable way nor our entire driving habits in a noticeable way.

        In fact I even said it doesn’t matter if people drive more. If you get more mpg you are saving. If you spend that saving on more driving that doesn’t ‘defeat the purpose’. You are getting more from the same amount and that’s a better state of affairs. Now I’m skeptical that people who got more mpg C4C cars really are driving so much more that they offset the improved mpg….but again why is that a problem?

        Report

        • @Boonton, Jaybird implied that those who got more efficient vehicles in C4C offset their improved mpg with more driving.

          Was there a dip in gas purchases in the weeks that followed C4C?

          If there was no reduction in gas purchases, I think that we have to assume an increase in driving, don’t we?

          Report

          • @Jaybird, for what it’s worth, I think that there may be an increase in driving with increased fuel economy, but I just can’t imagine it’s 1:1. I don’t believe that people are that conscientious of gallons used and more to the point a whole lot of driving is non-discretionary.

            Even if you look at annual miles driven in a small car vs. miles driven in larger cars, that can be misleading, too. When I had an 80-mile round trip commute, I was looking at getting as fuel-efficient car as made sense. Now that I live in a place where I can get across town in five minutes, fuel efficiency has fallen off my radar in importance. Cause and effect running in the other direction, I get the fuel-efficient car because I need to drive further and not so that I can drive further.

            Report

            • @Trumwill, I agree that there probably is an increase. I don’t think people think about it but it becomes a ‘gut feeling’. If you’re hitting the gas station a lot, that becomes a little bit of a drag on the idea of going out in the car. If all in the sudden you’re not hitting it as often, you think about gas less and will use the car more. I think it’s relatively subtle though and impacts driving at the margin. I don’t think you’re going to change your job commute to add many more miles.

              Report

        • @Boonton,

          It seems to me that you are conceding, then, that one of the main so-called environmental benefits of the program was not even measurable as such — and that it was more of a private benefit, like me being forced to paint my rich neighbor’s house, while leaving the rest of the neighborhood shabby.

          In other words, it was in no sense a benefit the government has any business providing.

          Report

          • @Jason Kuznicki, Well actually we did measure it. The DOT knows the average mpg of cars turned in and the average mpg of cars that were purchased by the program. From that we know that the 700,000 vehciles in the US fleet have improved mpg.

            Now we don’t all have secret gps chips implanted in our cars reporting how many miles we drive. I have seen sites that say we average 15,000 miles per year. At that rate, the average car turned in would have burned 949 gallons a year and the average replacement car would burn 602 gallons a year for a savings of 347 gallons.

            Now some critics assert that when people get higher mpg cars they drive as much as 25% more. If that happens then the average C4C car would drive 18750 miles in a year and burn 753 gallons. Still a savings of 196 gallons a year.

            In terms of private versus public benefits, I’d say the physical cost of the $4500 was stimulus and at least offset itself with the multiplier. The public cost then comes down to the destroyed capital. The benefit then is split between the private owner’s gas savings (let’s say $1000 a year from 300 gallons $3 a gallon) and whatever the non-market costs of oil are (if you say $0.50 a gallon that’s $150 a year). So total benefit per year is something like $1150. I don’ t think it could be much lower (gas is nearly $3 a gallon) but it might be higher (the multiplier from stimulus is larger when you’re deep in a recession, the non-market costs of oil could very well be a lot more than $0.50 a gal).

            You’re right, it can get hard to measure but that’s not much support for calling it a ‘failure’. To use the analogy of your house, yes you increase the value of your home by painting it, by mowing the lawn, keep the yard clean etc. No you can’t really measure that in the sense that you can say to yourself every time you mow the lawn you’re increasing your home’s value by $75.23. You can make some rough estimates, though, so you can figure out if you’ll get a bigger return on, say, a new paint job versus adding a level to your deck but they are only estimates.

            Report

          • @Jason Kuznicki, I think there’s a point to be made about accounting here. When a person switches from a low mpg car to a high one they produce savings in the sense that they use less gas per 1000 miles driven (namely about 23 gallons less per 1000 miles driven).

            That savings can be spent anywhere. A person can, for example, take a vacation on the savings. They can use it to jet ski, they can use it to eat out once a week and so on. They can even use it to drive more.

            IMO what you do with a savings doesn’t cancel out the fact that it’s a savings. If a person drives 18,000 miles a year instead of 15,000 because they have a higher mpg car that doesn’t neutralize the savings of the car. Presumably out of all the options available, the person has a need or desire for driving over things like taking a vacation. From an environmental perspective it’s not even very clear opting for more driving is all that bad a choice. Airplanes consume a huge amount of fuel yet if the 700,000 C4C drivers took trips to the Bahamas you wouldn’t criticize that because it would be almost impossible to know. But if they drive 10% more because they like their new, efficient cars, that gets jumped on as a ‘failure’.

            Report

  19. If there was no reduction in gas purchases, I think that we have to assume an increase in driving, don’t we?

    Feel free to produce the data and let us know what the results are. I’d be curious to know why you think you’d be able to measure the impact C4C against everything else. Why, for example, are the ‘few weeks’ after C4C so important? When do you put the most pleasure miles on a new car? When you first buy one or months later?

    I’d think the better way to study that would be to try to sample C4C owners but even that would be difficult.

    Report

  20. Also as I pointed out with the NY Times link, total miles driven in the US has fallen in 2009 and fallen even more in 2010. Even if you found gallons purchased by week you’re going to have to figure out what portion a change in 0.65% of the US fleet getting better mpg is compared to the overall downward trend. Even if C4C drivers offset their better mpg with more driving I don’t think you’d be able to see it even if you got data on gallons of gas sold per day by individual gas station.

    Report

      • @Jaybird, Actually we don’t really have ‘wealth destroyed by program’. We can only guess that it’s less than $4500 per car in most cases, probably a lot less. Someone looking for a good economics thesis can probably come up with something more solid by asking for the DOT for a listing of vehicles turned in by make, model, year and mileage.

        Report

        • @Boonton, if a car engine that is perfectly good (defined all loosey-goosey as “there are people out there willing to purchase this car and see it as a step up from what they are driving now”) is destroyed, then wealth is destroyed.

          Money is not wealth. Money is a convenient fiction/collective illusion/etc.

          Wealth *WAS* destroyed by this program, Boonton.

          Inexplicably.

          Report

          • @Jaybird,

            Money is not wealth. Money is a convenient fiction/collective illusion/etc

            Knock it off. Money is an illusion? Feel free to send me all the illusion in your bank account then !

            Wealth is not some infinite ‘feely’ thing. Yes a working car engine is an asset. It has value and that value is brought and sold in the marketplace. The value of the wealth destroyed could not be more than $4500 a car unless you have evidence of mass insanity on the part of C4C buyers. The value of the wealth destroyed was almost certainly much less than $4500 a car since there would be no point in going thru the C4C hassel for just a few dollars profit.

            Report

            • @Boonton, I don’t need mass insanity on the part of the buyers.

              I just need mass insanity on the part of the committees who created the bill.

              Like, for example, I picked this car out of the list of eligible cars:

              2001 Dodge Ram Pickup 2500

              I then put it into the Google and got some pricing info:

              edmunds.com/dodge/rampickup2500/2001/index.html

              Check it out.

              Report

  21. Like, for example, I picked this car out of the list of eligible cars:

    2001….

    I think you’re misreading stuff. 2001 is not a new car. Are you perhaps looking at the cars turned in under the program versus the ones purchased?

    Report

    • @Boonton, I was looking at the cars turned in.

      This car now trades for 6 grand (check the link).

      This is one of the cars that would have been destroyed if traded in (and it could well have been as it was one of the “full $4500!” cars).

      Wealth was destroyed.

      Additionally, it looks like while an additional 360,000 cars were, in fact, sold during the two months of cash for clunkers, sales following were off by about that much.

      Saying “it achieved its goals!” is looking like less and less of a defense of how successful it was. If its goals were downright pedestrian and backwards, it’s not a compliment to say that it achieved its goals.

      It accelerated sales (that were going to happen anyway) by two months at enormous cost and then destroyed a number of vehicles (360,000?) that are trading in the mid-four figures now… vehicles that, I reckon, get better gas mileage and have fewer problems than some of the pieces of crap that I drive past when I happen to drive through the non-middle class portion of town.

      Report

  22. Lost two replies, Kelly’s Blue Book says a 2001 Ram in fair condition is worth about $1900. That’s with 110K or so miles on it. Obviously more mileage or more dents and damage would mean even less value. And it’s not easy to actually get Blue Book value when you’re selling a used car.

    Are you seriously telling me that a large number of people decided to cash in a Ram they could have sold for $5-$6K in exchange for a $4500 voucher? Even if they didn’t want the hassel of selling a car privately I’m sure a dealer would have been happy to have brought such a car for $4500 rather than churning the paperwork and lines required for the C4C program.

    You are just all over the board here and not even paying attention to facts. You assume just because a 2001 Dodge Ram was eligeable to be turned in under the program that lots of such cars were, in fact, turned in. The Dodge Ram isn’t even on the list of the top ten trade ins. You’ve previously asserted that the top trade in was an SUV for an SUV making it an ‘SUV upgrade program’….despite the fact that 81% of the trade ins were trucks in exchange for cars. Must I really supply all the facts here while you just make stuff up?

    Report

    • @Boonton, No, Boonton. I’m not saying that it would have been traded in for a mere $4500 if it could have been traded in for $6000.

      So why did the government include it in the list of cars eligible?

      edmunds.com/cash-for-clunkers/eligible-vehicles.html

      Is the government insane?

      I’m using Edmunds because, hey, it’s what shows up first in google.

      According to CNNMoney, the #1 vehicle was, in fact, the focus… but the #2?
      The Escape.
      The #4?
      The Ford F-150.
      #7? The CR-V.
      #8? The Silverado.

      Check it out here:
      money.cnn.com/galleries/2009/autos/0908/gallery.clunker_top_10

      Now, saying it was an “SUV for SUV” was an overstatement… but 4 of the top 10 cars traded for were either trucks or SUVs.

      Given that sales were down in the months following by the amount that they were up in the months during ( npr.org/blogs/money/2010/09/02/129608251/cash-for-clunkers ) makes this less and less of a “success”, Boonton.

      Even if it did meet its ostensible “goals”.

      Report

      • @Jaybird, So why did the government include it in the list of cars eligible?

        Presumably because it gets low mpg. More importantly how do you logically go from ‘on the list of vehicles eligeable’ to be traded in to it being a major portion of 360,000 vehicles traded in?

        Your hysterics on the RAM makes no sense at all. Why shouldn’t it be on the list? It’s a truck that has a very low mpg, why wouldn’t we want people to trade it up? Your reasons seem to be because you think the 2001 RAM is worth more than $4500. So what?

        1. Are you saying the owners of RAMs are stupid and will sell for $4500 when their trucks are really worth $6000? If so then you have a big argument against the market economy. If people are so stupid that they can’t make even basic decisions about how much their stuff is worth then what hope is there for markets?

        2. Do you buy used cars sight unseen with nothing but quotes from Edmunds as though used cars were shares on the S&P 500? You do realize that just because the ‘average’ 2001 Dodge Ram is worth $5000 that doesn’t mean every 2001 Ram is worth $5000. Some are in exceptional shape and are worth more. Others are worth way less.

        Now, saying it was an “SUV for SUV” was an overstatement… but 4 of the top 10 cars traded for were either trucks or SUVs.

        Well we have the average mpg of both the traded in vehicles and the new vehicles and there’s certainly a serious improvement.

        http://www.dot.gov/affairs/2009/dot13309.htm has the categories and unfortunately SUV isn’t a category but 404K vehicles brought were cars, 231K were ‘category 1 trucks’. You can see on the trade in chart most of the trade ins were category 1 and 2 trucks While there might have been some SUV For SUV trade ins they were wither for improvements in mpg or were swamped by other trades that yielded much better mpg savings.

        On the issue of sales being down:

        We have the wsj chart already of the annual rate of auto sales. While sales did track down for a month or two following C4C they are higher in the half year after the program than the half year before. I’m not clear, though, why this is such a horrible thing to you. Moving some future demand forward to the present moment doesn’t seem like a bad idea when you’re in the deepest part of a recession that’s bordering on depression. I’ve asked you before what is the ‘cost’ of this and you’ve been unable to really articulate any cost. I can see in a period of full employment this could be a cost by misallocating labor to handling the spurt of sales and then losing it when the ditch comes but in 2009 we had plenty of labor to spare so what’s the cost?

        Report

        • @Boonton, More importantly how do you logically go from ‘on the list of vehicles eligeable’ to be traded in to it being a major portion of 360,000 vehicles traded in?

          Did I say that it was?
          Or are you just implying that I said that it was in an effort to make my argument look weaker?

          What I said was this (and I will quote myself): “I don’t need mass insanity on the part of the buyers. I just need mass insanity on the part of the committees who created the bill.”

          I then pointed out one of the cars on the list.

          I then said, and I’m quoting myself again, “This is one of the cars that would have been destroyed if traded in (and it could well have been as it was one of the “full $4500!” cars).”

          Nowhere did I say that it was a major portion of the 360,000 vehicles traded in.

          The fact that you think that I said this may go a long way to explaining how you don’t understand my argument…

          While sales did track down for a month or two following C4C they are higher in the half year after the program than the half year before. I’m not clear, though, why this is such a horrible thing to you.

          It’s not particularly horrible if it happens as part of the ebb and flow of normal activity.

          If, however, the government spends $24000 per car sold during cash for clunkers, I do have grounds to say something to the effect of “No, Boonton, the program was not ‘successful’.”

          Doing nothing at all would have done less damage.

          Which has been my point since the beginning.

          Report

          • @Jaybird,
            If, however, the government spends $24000 per car sold during cash for clunkers, I do have grounds to say something to the effect of “No, Boonton, the program was not ‘successful’.”

            As demonstrated, this figure means almost nothing. You might as well say the gov’t spent $750,000 per blue car with white strip purchased.

            Also since its been asserted we can’t count stuff we can’t precisely measure then you can’t count ‘cars that would have been brought anyway’.

            What I said was this (and I will quote myself): “I don’t need mass insanity on the part of the buyers. I just need mass insanity on the part of the committees who created the bill.”

            So if the gov’t doesn’t prohibit someone from making a foolish transaction that’s all that’s needed for foolish transactions? Foolish transactions aren’t prevented by, ohhh, say the fact that people are typically interested in getting the best deal they can?

            Nowhere did I say that it was a major portion of the 360,000 vehicles traded in.

            Hmmmm….

            It accelerated sales (that were going to happen anyway) by two months at enormous cost and then destroyed a number of vehicles (360,000?) that are trading in the mid-four figures now…

            So again how do you logically go from it being theoretically possible for someone to turn in a vehicle that trades in the mid four figures to a number of vehicles (360,000?) being worth more than $4500?

            Report

            • @Boonton, As demonstrated, this figure means almost nothing. You might as well say the gov’t spent $750,000 per blue car with white strip purchased.

              Except while there are ways to make the amount per car go up (only red cars! only priuses!), there isn’t any way to make this number go *DOWN*.

              Because you cannot make the number of cars sold using the program go up any more than they did… and you can’t make the cost of the program go down.

              $24000 per car is as low as that ratio is going to get.

              And it’s still an awful, awful ratio considering what we got out of it (specifically, an acceleration of purchases that were likely to have been made anyway rather than an incentive to purchase something that would not have been purchased otherwise).

              Foolish transactions aren’t prevented by, ohhh, say the fact that people are typically interested in getting the best deal they can?

              Of course they are.

              That is the point of my entire argument. People are rational, for the most part. The program made a bunch of rational people who were going to make a particular trade anyway hurry up their decision-making process by a couple of months.

              And this cost the government $24,000 per car sold.

              So again how do you logically go from it being theoretically possible for someone to turn in a vehicle that trades in the mid four figures to a number of vehicles (360,000?) being worth more than $4500?

              I didn’t say “more than $4500”. But let’s say that mid-four figures goes from $3500-6500.

              Would you like me to go edmunds?

              Well, the #1 car on the list of cars traded in was the 1998 Ford Explorer.

              On Edmunds now, prices range from $2351 to $3663.
              #2 was the 1997 one. $2132 to $3331
              #3 was 1996 Ford Explorer. $1823 to $2929.
              #4 was the 1999 Ford Explorer: $2626 to $4145.
              #5 is the Jeep Grand Cherokee… hrm, no year listed. I’ll take the bottom price from 1996 to the best price from 1999. $2410 to $5576.
              #6 is the Jeep Cherokee. No year given, same deal. $1454 to $4076.
              #7 is the 1995 Ford Explorer $1459 to $2613
              #8 is the 1994 Ford Explorer (dang, lotta Ford Explorer action in this list!) $1093 to $2096.
              #9 is the Ford Windstar $1530 to $1745
              And last, but not least! #10.
              The 1999 Dodge Caravan. $2153 to $3079.

              How many of those are dancing around the definition? Four of them? Including the #1 pick? Eh, if you want to argue that they aren’t in the “mid” four figure range but “mid to low” four figure range, I wouldn’t argue overly.

              Seems odd that an incentive of, at the very worst, less than $3500 would cost $24000, no?

              Did the government get its money’s worth?

              I would think not.

              I very much don’t see how the program could be considered “successful” except by people who are ignoring its cost.

              Which seems like quite the camel to swallow.

              Report

    • @Boonton,

      The value of the wealth destroyed was almost certainly much less than $4500 a car …

      But it was NOT zero. That’s the point.

      Thought experiment: Let’s run the program the exact same way — except leave the cars where they are, and let the owners sell them.

      We then skim a little tiny bit of money off the top of the program and give it to the Sierra Club, to compensate for the environmental damage, which both sides in the discussion now admit was on net very small.

      Altogether, we’d have a more efficient program, which tells us that something is still badly amiss. You’re still defending a broken window, and it’s getting more and more embarrassing as we go.

      Report

      • @Jason Kuznicki, 1. I’m not sure what the ‘environmental damage’ is of adding higher than average mpg cars to the US fleet.

        2. I never said the cost of capital destroyed was not zero. I’ve said that multiple times which is why I’ve pointed out saving 300 gallons or so per year pays for much of the capital cost within a year or so (assuming car owners are not as stupid as Jaybird’s impression of 2001 Ram owners). If I ever said there was no capital cost then there would be no need to estimate the ‘payback’ of the cost from the gas savings.

        3. Altogether, we’d have a more efficient program, which tells us that something is still badly amiss. You’re still defending a broken window, and it’s getting more and more embarrassing as we go.

        What’s embarrassing is that you guys make up all your facts while I do the leg work here. Here is a partial list of failed assertions that wouldn’t have been made if you guys actually put some effort into researching what you’re talking about:

        * The cost per car is in the six figure range.
        * Used car prices are now high because the program destroyed a portion of used car supply.
        * The typical trade was SUV for SUV
        * The 2001 Dodge RAM made up nearly half the trade ins.
        * That owners inexplicably traded in their cars for less than they were worth.
        * The value of the cars destroyed was over $5000.
        * A used car engine of a low mpg car is of such special value that it cannot be expressed in terms of crass money!
        * That there was a huge crash in auto sales post program.
        * That slight shifting of demand (about 11% of yearly sales) from a few months in the future to the present is a “huge cost”.

        The assertion here wasn’t that the program could have been better, that it could have been more efficient but that it was a FAILURE. After spinning and respinning a dozen failed claims the only thing you have to fall back on is what I’ve said all along, that destroying the used cars was a capital cost.

        What’s annoying here is that maybe if you guys are willing to give up your thrice refuted red herrings we can actually address that one valid concern, which I already did. Namely….

        In a perfectly efficient market gas prices would reflect their full costs, which is more than the price you see at the pump. Low mpg cars would be less valuable in such a perfect market which means the bottom end of the used car supply would end up in the junk yard sooner. Since the program was small and only impacted a small portion of the fleet (note my comments on how I wouldn’t feel as comfortable if the program was $30B or $300B as opposed to $3B), I suspect a lot of the capital destruction was capital that should have been destroyed anyway. Of course, my ideal preference is to try to estabish a more transparant market removing subsidies, hidden costs and other distortions accross the board but that not being the world we’re going to get this policy IMO probably offset in the opposite direction other foolish policies making it actually a zero cost policy.

        Report

        • @Boonton, my point about the 2001 Ram was not that 2001 Ram owners were stupid, but that the government was.

          It remains that.

          Though I’m considering expanding that to “defenders of cash for clunkers”. Thanks for keeping me posted as to whether or not I should!

          Report

        • @Boonton, The 2001 Dodge RAM made up nearly half the trade ins.

          Where was this asserted, Boonton?

          I’m guessing you’re either deluded and cannot find such an assertion *OR* you are deliberately lying and cannot find such an assertion.

          Could you source the assertion?

          All you have to do is scroll up, cut, and paste.

          We’ll wait here.

          Report

          • @Jaybird,

            As previously quoted:

            It accelerated sales (that were going to happen anyway) by two months at enormous cost and then destroyed a number of vehicles (360,000?) that are trading in the mid-four figures now…

            The only car you cited that you think is worth mod-four figures is the 2001 Ram but yes it is a leap to think you meant only the 01 Ram made up the 360K ‘mid-four’ cars you were talking about.

            Report

            • @Boonton, the car I cited that edmunds says is in the mid-four figures is the 2001 Ram.

              Would you like me to go through Edmunds lists and find more cars?

              I could, if you want.

              I just picked that one because it was one of the most recent cars that struck me as most likely to be worth a good deal more than Cash4Clunkers was offering (and, as such, one of the vehicles that would have been preferable to the drivers of a 1977 Datsun Pickup and would get better gas mileage than said pickup, have better emissions, so on and so forth in an effort to explain how, no, wealth was, in fact, destroyed by this program).

              Report

        • @Boonton,

          So let’s take on each of these one at a time:

          *The cost per car is in the six figure range.

          I don’t see anywhere above where such was asserted.

          I do see where it was pointed out that the cost per car was $24,000.

          I can source that number to here:
          money.cnn.com/2009/10/28/autos/clunkers_analysis/index.htm

          Your assertion that the cost per car was in the six figure range is not one that I see above (but maybe I missed it).

          Could you point out where such was said?

          * Used car prices are now high because the program destroyed a portion of used car supply.

          It was pointed out that “This may explain why used car prices are higher.”

          Without getting into the semantic difference between “high” and “higher”, I’ll just say that your beef here is with the person who left the comment at 09.06.10 at 10:34 am.

          * The typical trade was SUV for SUV

          This was conceded to be overstatement but it was also pointed out (and sourced) that 4 of the top 10 vehicles traded for were either trucks or SUVs.

          * The 2001 Dodge RAM made up nearly half the trade ins.

          This does not appear anywhere, that I can see. Could you source this assertion? Scroll up and find it, please.

          * That owners inexplicably traded in their cars for less than they were worth.

          This assertion was also not made anywhere that I can see. Could you source it? The closest I could find is the argument that the government was crazy… not the consumers. The inclusion of cars worth more than $4500 was given as evidence for this.

          * The value of the cars destroyed was over $5000.

          I don’t see where this was asserted either. I do see where the (trivially true) argument was given that *IF* this car was traded in, *THEN* it would have been destroyed and, since it trades, according to Edwards, at more than $4500 then wealth would have been destroyed by this program.

          * A used car engine of a low mpg car is of such special value that it cannot be expressed in terms of crass money!

          Where was this made? This seems similar to your arguments, rather than the arguments of the glibertarians on the board.

          * That there was a huge crash in auto sales post program.

          It was pointed out that there was a huge boost in sales for the program and then sales plunged.

          Do you dispute the facts cited here: money.cnn.com/2009/10/01/news/companies/autosales/index.htm

          Or is this a “you said ‘crash’ when you should have said ‘went down sharply'” issue? Is phrasing the problem here?

          * That slight shifting of demand (about 11% of yearly sales) from a few months in the future to the present is a “huge cost”.

          This slight shifting of demand from a few months in the future wasn’t only the assertion… the fact that it cost $24000 per car was the huge cost and it’s certainly a huge cost (a qualitative judgment) compared to many other plans that could be considered (including, of course, the free “doing nothing” plan).

          In any case, I look forward to you sourcing the claims you claim were made that I wasn’t able to find (I helpfully said which ones I was not able to find being made).

          The others? Well, they seem to be problems with phrasing or very small beer indeed.

          Report

          • @Jaybird,1. Six versus five figures

            – Granted the claim was five figures. Nonetheless as the program’s goals were only partially about selling additional cars this calculation is as meaningless as saying the cost per blue car with white strip sold is $250,000.

            2. Used car prices higher

            81% of vehicles traded in were trucks. So not only was a tiny portion of the US fleet traded in but only a tiny portion of that wsa cars to begin with. As asserted by the author of this post:

            And the payback isn’t long in coming — today’s used car prices are soaring owing to reduced supply.

            Sorry ‘soaring used car prices’ were asserted to be attributable to C4C and that’s not plausible if you bother to give the facts even a casual glance.

            3. “I’ll just say that your beef here is with the person who left the comment at 09.06.10 at 10:34 am.”

            I didn’t say all the silly assertions belong to you only.

            4. SUV for SUV / 4 of the top ten cars purchased were SUVs

            This you conceded but continue to backtrack. The top 10 vehicles purchased were:

            Toyota Corolla
            Honda Civic
            Toyota Camry
            Ford Focus FWD
            Hyundai Elantra
            Nissan Versa
            Toyota Prius
            Honda Accord
            Honda Fit
            Ford Escape FWD

            Of these only the Escape jumps out as possibly an SUV to me (seems to be a small one though). I didn’t look up all of them so I could be wrong. Maybe the Prius and Versa are SUVs

            5. “The 2001 Dodge RAM made up nearly half the trade ins.”

            I quoted you twice now, look at the previous comments. You implied 360K of the vehicles were worth in the mid four range. Granted you didn’t say the RAM made them up.

            6. “That owners inexplicably traded in their cars for less than they were worth.”

            The only way for valuable used cars to get destroyed by the program is for owners to trade them in. Simply being able to trade in a Cadelliac Escalade worth $20K for $4.5K doesn’t mean thousands of such Caddys were destroyed by the program.

            7. That used car engines can’t be measured by money? “Where was this made? This seems similar to your arguments, rather than the arguments of the glibertarians on the board.”

            Money is not wealth. Money is a convenient fiction/collective illusion/etc.

            8. “It was pointed out that there was a huge boost in sales for the program and then sales plunged.”

            Yet for the whole year sales tracked higher post-program than pre-program.

            9. “This slight shifting of demand from a few months in the future wasn’t only the assertion… …”

            No but you did assert it was a cost. How is it a cost for about 11% of sales or so to shift from the future to the present during a period of slack demand and economic crises?

            Report

        • @Boonton,

          Please reread what I wrote. I was saying that we would need to compensate, in our hypothetical program — which was different from the real-world one — for the environmental damage caused by leaving the clunkers on the road.

          This compensation can be easily done without destroying the clunkers’ economic value. And it should have been done, in a more efficient program, even if we grant the legitimacy of all the real-world program’s goals. On its own terms, C4C was economically inefficient.

          Put it another way: Suppose we had Cash for Clunkers exactly as it was, plus we broke a window.

          Good idea, or no? Obviously no. It’s worse than the program as it was. So let’s not break those cars either. Let’s repurpose them, at the very worst.

          As to your points:

          * The cost per car is in the six figure range.

          Never claimed it.

          * Used car prices are now high because the program destroyed a portion of used car supply.

          Did claim it, though only as one factor among many.

          * The typical trade was SUV for SUV

          * The 2001 Dodge RAM made up nearly half the trade ins.

          Apparently both are pure fabrications on your part.

          * That owners inexplicably traded in their cars for less than they were worth.

          Certainly they didn’t. Again, you’re not following the argument. If someone gave me a cash incentive to trade in my car, I’d do it, provided the incentive was greater than the transaction costs. This isn’t hard to understand. Free money is always awesome from the recipient’s standpoint, even if it doesn’t do anything great for the overall economy. (The latter incidentally is what I’m saying.)

          * The value of the cars destroyed was over $5000.

          Insufficient data. Mighta been. I don’t know.

          * A used car engine of a low mpg car is of such special value that it cannot be expressed in terms of crass money!

          This would appear to be your side’s claim, not mine.

          * That there was a huge crash in auto sales post program.

          There was a significant drop in sales after the program ended, yes.

          * That slight shifting of demand (about 11% of yearly sales) from a few months in the future to the present is a “huge cost”.

          No, we’re only claiming that cutting one end off a rope and tying it onto the other doesn’t make the rope any longer.

          Report

          • @Jason Kuznicki, Jason Kuznicki

            Please reread what I wrote. I was saying that we would need to compensate, in our hypothetical program — which was different from the real-world one — for the environmental damage caused by leaving the clunkers on the road.

            I’m not really sure I follow this. I suspect the destruction of used cars wasn’t really a broken window. They were cars and trucks that would have been taken off the road IF other policies made the price of fuel more accurately reflect its full environmental and political costs. In this sense the policy of destroying the C4C cars actually was offsetting other policies that cut in the other direction.

            Consider a policy that puts a slight tax on corn syrup. On its own its a tax with various distortionary effects but in the larger context of the way things actually are corn receives huge benefits from foolish subsidies. A corn syrup tax would slightly offset some of that. Of course the ideal policy in the ideal world would be no syrup tax and no subsidies.

            Put it another way: Suppose we had Cash for Clunkers exactly as it was, plus we broke a window.

            Good idea, or no? Obviously no.

            Or do you mean to say Cash for Clunkers without breaking a window because the major criticism seems to be destroying the used cars was destroying useful capital? I’d say bad idea because destroying the cars actually increased the benefits of the program for the scale that it was enacted for. If C4C was, say, ten times larger I’d say you’d start destroying capital that really was useful. At $30B or $300B I’d probably say C4C wouldn’t be as good but at $3B I think it might have hit the sweet spot of ‘just right size at right time’.

            For the stuff you didn’t claim, I meant it to be a list of all the stuff tossed out so far by everyone. There’s no fabrications there but probably some misunderstandings. Jaybird did claim, for example, that a number of cars traded in were very valuable and the only basis was the 2001 Dodge Ram was on the list as an acceptable car to trade in.

            This is an important point because if owners did trade in cars that were very valuable that would increase the value of the capital destroyed possibly tipping it away from simply correcting a market inefficiency and towards destroying useful capital. But the only way to make that argument viable is to have owners acting very irrationally

            Report

            • @Boonton, Jaybird did claim, for example, that a number of cars traded in were very valuable and the only basis was the 2001 Dodge Ram was on the list as an acceptable car to trade in.

              I did?

              Could you point me to where I said this? Cut/paste, please.

              (If you want to point out where I said it was $24000/car, feel free to not bother and allow me to say “good one”.)

              Report

            • @Boonton, Actually a program that paid for improved windows on homes probably would destroy the original windows, I never heard of a market for used windows. So there too you’d have to confront the fact that you are destroying some capital that could, in theory, still give you some more years of useful service.

              The $24000 number remains pretty stupid for reasons we’ve been over a thousand times now.

              Was it wisdom that caused the program to his a ‘sweet spot’? I don’t think so. It was the fact that it was small (small steps are less likely to result in you breaking something), that it was fast (hanging around a long time gives people the chance to figure out how to game the system), and that in a depression just about anything you spend money on returns its value in stimulus.

              Let’s just pretend that every car was in fact a ‘new sale’ that wouldn’t have happened without the program (not all that crazy, it just requires assuming very elastic demand for cars). The ‘cost per car’ then would be just south of $4500. I don’t see why that would make such an important difference to Jaybird. 125000 new sales or 700,000 new sales is important for the auto industry but for the economy it simply means that instead of other things (movies, books, plastic toys at Wal-Mart, carpets), cars were purchased. If your interest is the entire economy why is motivating new car sales by itself such an important metric for you?

              Report

            • @Boonton,

              As I’ve said again and again, I don’t agree that the program hit any sort of sweet spot. You give people free money, and they will like you. But I have never seen you address the obvious improvements to the program that might have been made, even granting that stimulus spending in a recession is wise (which I’ve been doing all along for the sake of having an interesting argument, although it’s not a premise I’d ordinarily grant).

              Again: Why not give the cars to the needy? Yes, there would be some negative environmental impacts in doing it. So you spend a little more on cleanup, or solar, or whatever — and you’ve got, on balance, a more efficient program that does every single thing you wanted. Without destroying any cars or engines.

              It’s that, or we could go around breaking windows.

              And beyond this very simple alternative, I don’t see that we’re making any headway, despite weeks of discussion. So barring any really surprising changes of heart, I’m probably going to bow out here.

              Report

  23. Jaybird

    @Boonton, my point about the 2001 Ram was not that 2001 Ram owners were stupid, but that the government was.

    Why? The only one looking stupid here is you.

    1. You ignore the fact that Kelly quotes a fair condition 2001 RAM as worth $1900, nowhere near $4500.
    2. You ingore the fact that its often very hard to achieve a ‘blue book’ price for a used car.
    3. You ignore the fact that even if the average 2001 RAM is worth $5000 or more that’s just an average. Some used RAMS are worth a lot less.

    The only reason for asserting including the RAM is stupid is assuming RAM owners are too stupid to realize their trucks are worth more than $4500 and therefore the gov’t must protect them from making a bad deal.

    Report

    • @Boonton, I see that you say that Kelly says that… but the only sourced assertion I see is the one that I provided.

      Go ahead. Cut and paste that link into your browser.

      If you have a different link that argues differently, *PLEASE PROVIDE IT*.

      I gave my take and provided my sources. What more do you want?

      As for the cars that were dying, if not dead, and shambling in on their last legs into the dealership… well, I kinda think that those cars would have been off the road rather soon anyway so touting the benefits of cash for clunkers by getting a soon-to-be-dead car off of the road is disingenuous at best.

      At this point, you seem to be praising a program that subsidizes people who don’t maintain their cars particularly well.

      I don’t see how that’s something to crow about.

      Report

      • @Jaybird, I’ve tried posting links twice now, but both times they got caught in moderation hell even though I swear I removed the httyouknowwhat. Anyhow, as far as the KBB value goes, he is right. The average trade-in for a fair-condition 108k/mile standard-everything RAM 150 was somewhere south of $2000. It was a little higher in the NADA directory. The NADA also supplied a number for what it’s clean retail value would be and that actually matched up with the Edmunds numbers you provided (around $6k). So I think the difference here is trade-in vs. retail.

        Unfortunately, if you want to see the numbers for yourself, you’re going to have to go to KBB’s or NADA’s website and fill out the form yourself.

        Report

        • @Trumwill, okay, fair enough. Here’s my point in response:

          Given that the price per car is $24000 (source: money.cnn.com/2009/10/28/autos/clunkers_analysis/index.htm) the difference between a car being worth two grand and four and a half-grand is miniscule when it comes to how much of a waste the program turned out to be.

          Let’s say that it would be smart for me to trade in my vehicle for the C4C trade-in value because, hey, free $2500 in trade… that’s smart on my part.

          How spectacularly dumb is it for the government to spend 21500 to get me an extra 2500?

          Report

          • @Jaybird, preach it, brother!

            Though I don’t buy the notion that C4C bears more than minimal responsibility for the increased in used car prices, it’s still dumb policy. The only benefit I’m sold on is emissions-reductions, and this is a really roundabout and inefficient way of going about that.

            Report

          • @Jaybird, So I take it you are now backing down on your assertion that the 2001 Ram is worth around $5K and you are backing down from your assertion that it was stupid for the gov’t to not exclude ‘valuable’ cars from the list of acceptable cars for trade in? You’ve spent a dozen posts wasting our time arguing something that you will now pretend you never said.

            Report

            • @Boonton, given that my point was never the 2001 Ram but that the list contained cars with a fair amount of life left that would be worth trading to people who could use the heck out of them (and a good measurement of that was “price”), I stand by my underlying point.

              Would you rather I use the 1999 Infiniti QX4? The 2003 Isuzu Rodeo? 2003 Isuzu Rodeo Sport? The 1999 Jaguar XJ-Series? (Seriously? There are Jaguars on this list??? What the hell???) The 1996 Lexus LX 450? The 1998 Mercedes-Benz M-Class? (They have Mercedes-Benz cars on this list…)

              The point wasn’t that it was a 2001 Ram, Boonton.

              Report

  24. Kellys is at Kbb dot com. For some reason I’ve tried topost the link multiple times but the site won’t let the comment up. Perhaps it it’s on a spam list?

    Anyway, there’s a difference between a low value used car and a car on its last legs. I’ve driven used cars that couldn’t fetch more than $500 for multiple years before hitting a repair bill making it better to junk the vehicle.

    Report

    • @Boonton, I’ve driven used cars that couldn’t fetch more than $500 for multiple years before hitting a repair bill making it better to junk the vehicle.

      Might be nice to have the option to swap one of those out with a 2001 Ram, no?

      Report

  25. Maybe its me but I’m having trouble keeping track of all the ‘replies’….I’m more used to comment threads that are sequential rather than the branching structure here so if no one objects I’m just going to throw my replies to the bottom:

    1. Might be nice to have the option to swap one of those out with a 2001 Ram, no?

    Yes but it wouldn’t have helped me. I believe the C4C program required you to have the clunker registered and insured for 6 months or a year beforehand. This was to prevent people from pulling cars out of the junkyard to claim the credit.

    2. You can’t make the $24,000 figure go down.

    Yea you can. How many of the people who were going to buy cars anyway opted to buy a car with better mpg than they otherwise would have? That alone would lower the $24K figure (assuming you got an accurate picture of how many ‘would have’ brought anyway).

    3. Incentive for people not to maintain their cars.

    This would be a valid problem if the program wasn’t temporary or if it was much larger than it was. Basically if you’d have to register and insure your $500 dying car a year before the program was passed. You would have to know ahead of time not only that this program was coming but also that your dying car wouldn’t die too soon leaving you carless before the program was passed. As it stands if you had a car that was about to die and you traded it in at C4C this is less of a subsidy and more of a stroke of luck for you.

    Report

    • @Boonton, it seems to me that the whole program was certainly improvable, no?

      As it stands, I think that the fact that in the middle of a crisis that *THIS* was the best that we could hope for from our Congress/Senate to get to the President to sign is indicative of a lot of problems.

      Moreover, I’m willing to say that if the Republicans wanted to do similar under Bush, we would have ended up with something similarly dog’s breakfasty. (Perhaps different groups would have been the beneficiaries of largess… but a reshuffled crap hand remains a crap hand).

      Report

      • @Jaybird,

        @Boonton, it seems to me that the whole program was certainly improvable, no?

        Yes, but then there’s not much in life you can say is ‘not improvable’. On the other hand it appears to have done good and it could have been a lot worse. For example, it would have been a lot worse if you had been allowed anywhere near the design of the program. Instead you’ve been safely relagated to backseat driver where the potential harm you can inflict is minimized.

        As it stands, I think that the fact that in the middle of a crisis that *THIS* was the best that we could hope for from our Congress/Senate to get to the President to sign is indicative of a lot of problems.

        Last I checked we have/had a $750B stimulus, $700B TARP, as well as massive stimulus from the Fed. This is a very tiny program relatively speaking that appears to have caused no real harm outside of your imagination.

        Report

        • @Boonton, “On the other hand it appears to have done good and it could have been a lot worse.”

          I don’t see how it appears to have done good.

          Did it boost sales? Yes.
          Did sales plunge after the program failed? Yes.
          Did the amount sales plunged more or less map to the amount sales improved?

          If the answer is yes then what we have done is nothing more than shuffled up sales by a couple of months. That’s it. Instead of buying something in November, they bought it in July.

          That’s *ALL* that happened.

          If we start looking at stuff like “well, maybe the improved sales had benefits!” we then get to compare those benefits to the costs.

          And the costs, as have been demonstrated, were $24000 per car. That’s at the low end. You can make that number go up (as has been demonstrated) but you can’t make it go down. You can’t make it so that more cars were sold because of the program. That number exists. You can’t make it so that the program didn’t cost as much as was spent. That number exists too.

          Was the good that we got worth the $24000 per car?

          I think it’s really easy to reach the conclusion that… no. It wasn’t. Certainly not to get a whole bunch of Ford Explorers off of the road.

          Report

  26. Did it boost sales? Yes.
    Did sales plunge after the program failed? Yes.
    Did the amount sales plunged more or less map to the amount sales improved?

    If the answer is yes then what we have done is nothing more than shuffled up sales by a couple of months.

    You already answered that question as you asserted $24K as the cost of each ‘true’ new sale generated. So you’ve already asserted that the program did not simply shift future sales to the present.

    Although you haven’t explained why ‘shifting’ should be considered a totally useless exercise during a period of crises.

    And the costs, as have been demonstrated, were $24000 per car. That’s at the low end. You can make that number go up (as has been demonstrated) but you can’t make it go down.

    Actually you can.

    1. Include the fact that the program almost certainly made some people buy cars with more mpg than they otherwise would have. This would lower the ‘cost per action motivated’

    2. As is often the case, estimates of what ‘would have happened’ are often based on averages of previous periods. This is all well and good provided you are looking at a time period that’s ‘typical’. If you’re not, however, your estimates are wrong. For example, if the estimate of 125K new sales is based on what auto sales typically look like in a recession, well you are not accounting for the pretty much undeniable fact that this is an exceptional recession. In that case there would have been much fewer sales if the program hadn’t happened which means your cost per sale is actually lower.

    3. As I pointed out the actual payments IMO were acting as pure stimulus in a depressed economy making their effective cost $0. The actual cost then is the value of destroyed capital, which you’ve tried to imply is very high but have provided no good explanation for why other than the possibility that 2001 Dodge Ram owners are exceptionally irrational and gov’t failed to save them from making a stupid trade.

    {Note again my ‘right size right time’ criteria. I wouldn’t necessarily make this claim if the program was done again during more normal times or if it was exceptionally larger than the $3B that was actually spent on it}

    Report

        • @Boonton, no, you didn’t show me. You asserted that it did despite the fact that we know exactly how much C4C cost and how many cars were purchased that registered as being covered by C4C and then we divide the numbers.

          I do not see how this metric is particularly useless when it comes to measuring what the program actually did.

          You’re just asserting that it is while, at the same time, arguing that the program was, of all things, *SUCCESSFUL*.

          Report

          • @Jaybird, Interesting link:
            http://www.carfax.com/car_buying/flood_damage_cars.cfx

            For those who argue now that C4C hasn’t caused the increase in used car prices but was a ‘factor’ (that’s a nice weasal word, technically your nephew ruining his 1995 Civic by driving it without oil is likewise a ‘factor’ in used car prices) might want to look at the above. The number of cars flooded each year totals tens of thousands. Large hurricans like Floyd, Ivan and Katrina have destroyed 75,000 to 100,00 cars at a clip.

            With that in mind we know that C4C destroyed 700,000 vehicles of which 84% were trucks leaving 16% cars or 112,000 cars. That basically means C4C had no more impact on used car supply than a year with a major hurricane.

            Why is the metric useless?

            Because, for the hundredth time, the program had multiple goals only one of which was to increase car sales. Judging it only by dividing estimated new car sales into the total $3B is as silly as dividing sales of blue cars into the $3B. The program served multiple goals:

            1. Get higher mpg cars on the road. This it clearly did.
            2. Stimulate the auto business. This it did
            3. Increase car sales. This it did.
            4. Stimulate the economy generally. This it did.

            If the goal was only to increase car sales, Congress could have just passed a first come first serve $4500 rebate for any new car purchase. Likewise when you have multiple goals you can have a success even if one goal isn’t accomplished. Even if, for example, there were ZERO net new car sales the program could still succeed by motivating higher mpg purchases than would have otherwise happened and moving demand from future months into the crises months.

            Report

  27. Jason Kuznicki

    1. Cars to the needy

    You need to be a bit more specific here. How would the needy be selected, what type of cars would they be given? How do you avoid the competition that would inevitably result in an Oprah style car giveaway? Such a program might be better than C4C, or it might be a diaster with Fox News profiling the spoiled 19 yr old kid who gets a brand new Prius because he qualifies as ‘poor’ since he has no income while his single mother neighbor making $22K a year driving a 10 yr old clunker doesn’t win an “Obama Car”.

    Just about anything can be done better in hindsight. Saying its possible to imagine a better progrma doesn’t make the one that really happened a failure and just imagining it is just that….your imagination. There’s no way to trust that brining your imaginary program from your head into real life wouldn’t reveal a host of problems that you are not anticipating. Even so saying program A happened but program B would have been better is not the same as saying program A failed. Jaybird’s assertion was that the program did more harm than good. To date I don’t think that’s been established.

    2. “It’s that, or we could go around breaking windows. ”

    Yawn, here we go again with the destroyed cars. First Jaybird tried to tell us they couldn’t be priced in terms of money. Then he told us lots of them were worth thousands. Look destroying the engines done because part of the goal was to increase the average mpg of the US fleet so clunkers were taken off the road in exchange for high mpg new cars. If, as Jaybird thinks, that was not one of the programs goals you could have simply droped the ‘destroy the cars’ part of the program and left everything else the same.

    But again here I think you have to consider that the program was relatively small, targetted mostly at the low end junk cars with low mpg, and that the market price of gas is lower than its true price. Putting all that together and it’s pretty hard to see much real cost from destroying the cars. I would suspect that most of the cars would have been destroyed if the market was more efficient. As I said if this was a $30B or $300B program the capital eaten up would more likely move beyond the junk that was just riding around because of accidential market inefficiencies and would start taking a bite out of real auto capital.

    Report

    • @Boonton,

      Cars to the needy… You need to be a bit more specific here.

      No, I don’t, and that’s just the beauty of it. If the goods are goods to someone, then that all by itself is more than enough to prove my point, namely that the program needlessly destroyed economic value. I don’t have to worry about distribution, because I already support a distribution system, namely the free market. I don’t need to propose any new system, let alone a perfect one, to point out inefficiencies in a given alternative.

      Your examples — the Oprah car giveaway and the case of undeserving recipients — only prove my point, which is that these cars remained economic goods. To someone. This was repeatedly pointed out at the time, and not merely with hindsight.

      As to your objections to Jaybird, they grow more and more fanciful with every post. I’d be peeved if I were him. But not one of your objections has anything to do with my points, which require only that the cars still had some non-zero economic value. The latter point you have lately underscored yourself.

      Report

  28. 1. The cars destroyed by the actual real life program probably did have a ZERO economic value for the reasons I’ve given multiple times. Yes they have value ‘to someone’ just as a field of growing corn has value to someone but at the same time we know if corn was properly priced (no subsidies, protectionism etc.) some of those ‘valuable’ fields would cease growing corn. The obsession with proving destruction of economic value here comes from the error of assuming a perfect market to begin with. Unfortunately the market for low end used cars is especially vulnerable to the distortions that we have built into our actual world as opposed to the textbok perfect one.

    2. Jaybird being peeved, nothing I said about his positions is false. He has staked out everyone one of them with only some trivial misunderstandings here (i.e. the ‘cost per car’ being 5 figures, not 6….a large number of cars destroyed being worth $4-$5K, not all 2001 Dodge Rams worth $4-5K).

    3. “. I don’t have to worry about distribution, because I already support a distribution system, namely the free market.”

    Yea ok but that’s not a hypothetical program to give 700,000 cars to the ‘needy’ (or if we are talking about keeping it to $3B let’s say giving away about 230K new cars worth $13K each). In terms of stimulus $3B is $3B but I suspect the C4C had a bit higher multiplier at work. Why? Because the program was structured to pull actual spending from individuals into the mix both in terms of those who opted to buy cars they otherwise weren’t (let’s say 125K) and from those who did either buying more expensive cars (to score higher mpg) or having $4200 more in cash that they could spend on other things. That plus going forward you have about $1K per year less spent on oil (which is 50% imported at this point?) that would likely be redirected towards other domestic spending.

    In terms of improving mpg? Assuming the new cars would have above average mpg I suppose it would push up the US fleet average, I suspect C4C did better by rewarding making a larger jump in mpg AND removing below average cars from the fleet.

    In terms of helping the auto industry? Jaybird’s ‘cost per real new car’ metric may come back to bite you here. Yes the gov’t buys 230K cars from GM to give away to poor people. Those poor people are either not going to buy new cars they otherwise were, are going to sell off used cars they already own or not buy used cars they otherwise would have. Somewhere in the market you’re going to see people who would have otherwise brought new cars not buy one because 230K people got free new cars from the gov’t. The actual net increase in auto sales, then, is still less than the number of vehicles involved.

    Report

    • @Boonton, Yes the gov’t buys 230K cars from GM to give away to poor people.

      How about not… but they, instead, take the 2001 Ram that the dude traded in for his new Prius and give that to some poor schmuck who is driving a 1977 Datsun pickup.

      Report

  29. I think we exhausted the ‘wasted capital’ argument. It’s either close to zero or some larger number that still isn’t that far from zero since most cars destroyed by the program almost certainly were worth far less than the $4500 voucher (even the 2001 Dodge Ram).

    I’d like to just talk about the last remaining argument here that has legs, namely $24,000 per new car sold.

    1. In terms of pure stimulus, $4500 is $4500. If the gov’t lowered payroll taxes by $3B for 700K people that would work out to about $4500 per person with no ‘action’. (Let’s forget about supply side arguments about people opting to take jobs if taxes are lower).

    2. So what if the program simply sent checks of $4500 to random people who brought new cars with high mpg’s a month or two befre? Certainly this program would be worse than $24K per new car sold in terms of Jaybird’s metric because zero new car sales would happen since the program simply rewards people who already purchased cars (OK technically some of them might take their $4500 to the dealer and buy another car but let’s forget about that). In terms of stimulus though, this program works….even though I think most of us would rather see #1 instead. #1 probably would help people who are struggling while this program helps people who are probably doing just fine….if you’re buying a new car in the middle of a depression you are probably relatively secure financially.

    3. Now imagine a program that paid $4500 only for sales that wouldn’t have happened. This would require some science fiction but just suppose a combination of data mining and one time sociological breakthru allows the gov’t to identify 700,000 clunker owners who are not going to buy a new car but would if offered the $4500 opportunity. In this case the cost per car sold works out to be $4500 since every car sold is a sale that wouldn’t have happened without the progrm.

    Per Jaybird’s metric #1 and #2 yield horrible results as they generate almost no new car sales. The cost per car sold is, let’s say, $3 Billion (let’s just say that 1 person opts to use his tax savings or surprise $4500 check to buy a new car). #3 is fantastic since its millions of times better than #1 and #2, it’s also about 5 times better than the real life C4C program (assuming $24K is correct).

    Thing is it doesn’t really look that much more fantastic. Yes its better for the auto industry but in terms of stimulus that’s a double sided coin. People who would have brought a car anyway had $4500 more to spend elsewhere, those industries wouldn’t get those sales if #3 was in effect. It might be better for our mpg goal. But more importantly Jaybird’s way of looking at things would say #3 is better than even #1….but #1 to me and most people looks best of all. Long story short his metric is measuring something real but using it as he does is crappy.

    Consider a slightly different way of using the metric. Suppose you’re head of Toyota Prius’s division and you want to boost sales. You lower the price by $4500. You were selling 575K units per year, but now you sell 700K units per year. All else equal, your ‘cost per new sale’ is not $4500 per unit. It’s more like $24,000 since you had to cut the price not only on the 125K additional units you sold but the 525K you were selling to begin with. $4500 of value ‘leaked’ to the 525K people who were buying Priuses anyway. This is true of most price cuts. Very few bring so many new customers to the mix that they are greater than existing customers. This doesn’t make it a bad policy, though, since it very well may be that the margin on the 125K additional sales offsets the $2,362,500 less in revenue you get from the original buyers (525K times $4500).

    What’s interesting about this though is:

    A. If your role was auto company executive, ‘cost per additional unit sold’ is a valid metric but still not one that really carries the day. The ‘cost’ might very well be $24,000 but the $4500 price cut may still make sense to the bottom line.

    B. In terms of the economy, the $4500 doesn’t ‘leak’. If Joe Smith brought a new car with the $4500 and his 7 brothers were going to buy a new car anyway but took their $4500 and redid their decks, went out to eat, paid off some credit card debts and so on. The stimulus is still valid. Toyota doesn’t get the $4500 the 525K customers saved (unless they buy other Toyota products with it). The economy though, does get it. So in those terms the $24,000 per car metric becomes even less valid.

    Report

    • @Boonton, Boonton, you are misrepresenting my argument.

      If I had merely said “Boonton thinks that Cash for Clunkers did great because $3 Billion is a small price to pay to temporarily move up car sales by a couple of months”, would you think that this would be giving your argument short shrift?

      Report

      • @Jaybird, Yes but it demonstrates the poverty of your argument. So C4C did something other than just make 125K additional net sales, it moved sales up a few months.

        So an alternative plan that simply made 125K new sales but did not move anything up a few months would do less than the actual C4C. But that same plan would ‘score’ the same by your metric, $24,000 per new car sold.

        See how your metric lacks helpfulness….just like calculating ‘cost per blue car sold’? We haven’t even added in yet the consideration that C4C drove higher mpg than would have otherwise happened.

        Report

        • @Boonton, the usefulness of the metric comes when you compare to “doing nothing”.

          What would “doing nothing” have done to car sales?

          Well, it would have spread the same amount over the same amount of time except done it evenly rather than front-loaded them.

          And how much would that have cost per car?

          Squat.

          How much would that have cost per blue car?

          Squat.

          How much would that have cost per Prius?

          Squat.

          How much would that have cost per blue Prius with Obama bumperstickers pre-applied?

          Squat.

          And how much did C4C cost per car?

          $24000.

          On a program that gave $4500 rebates on cars that probably would have traded in for a couple thousand less than $4500 that were then destroyed rather than fixed up and sold to people who would have seen them as a step up from what they were currently driving.

          Seems to me that “squat” is pretty defensible.

          Report

          • @Jaybird, What would “doing nothing” have done to car sales?

            Well by your source they would have been 125K less.

            Well, it would have spread the same amount over the same amount of time except done it evenly rather than front-loaded them.

            Except, though, front loading them was a good thing in this case.

            And still you dodge the benefit of having higher mpg cars than would have otherwise happened. Why not go to the ‘cost per blue car’ metric? If you want to be silly after having it demonstrated to you a hundred times go for the whole hog instead of half a one!

            Report

            • @Boonton, And still you dodge the benefit of having higher mpg cars than would have otherwise happened.

              Because the worst cars out there (those 1977 Datsuns that I keep complaining about) are still there.

              We’d get more benefit from getting rid of those than we would from getting rid of a number of Ford SUVs. Heck, we’d get a non-zero amount of benefit from straight up trading the Ford SUVs for those Datsuns.

              I said, at the time, that there were plans that I’d agree with… specifically ones that did more than shuffle sales around and didn’t destroy wealth.

              This plan was worse than doing nothing… thus I opposed it.

              Report

  30. Jaybird

    given that my point was never the 2001 Ram but that the list contained cars with a fair amount of life left that would be worth trading to people who could use the heck out of them (and a good measurement of that was “price”), I stand by my underlying point.

    Would you rather I use the 1999 Infiniti QX4? The 2003 Isuzu Rodeo? 2003 Isuzu Rodeo Sport? The 1999 Jaguar XJ-Series? (Seriously? There are Jaguars on this list??? What the hell???) The 1996 Lexus LX 450? The 1998 Mercedes-Benz M-Class? (They have Mercedes-Benz cars on this list…)

    Your underlying point remains refuted. What you’re missing, again, is (I’ll put the most important things on top).

    1. Car owners are unlikely to trade a valuable car in for less than its worth. Even if a ’99 Jauguar is on the list it’s unlikely anyone would turn one in for $4500 unless that was some exceptional mileage on it that made it worth a lot less than what it normally is. So screening the list for Blue Book values above $4500 would not have made the program any better.

    2. Blue book values are at best averages and tend to be ‘best cases’. Just because the book says so and so is worth $6,000 doesn’t mean it really is.

    3. Even if the book value is accurate it remains an average. A 2001 Dodge Ram owned by Hank Hill may very well be worth $5K because he pampers it like a baby. The same car owned by his neighbor might be worth only $500. Why should the gov’t exclude a low mpg car from the list just because some of those cars might be valuable?

    4. Even if a car owner was stupid and shows up with a car worth $10K seeking the C4C trade in, the dealer would find it faster and easier to just offer the guy $4500 for the trade in and avoid the C4C paperwork.

    5. This is last for everyone else but probably should be first for you. You asserted a large number of C4C vehicles were valuable. You based this on nothing other than the fact that because potentially valuable cars were on the list of cars acceptable for the program those were the cars actually turned in and destroyed. Your point remains that if gov’t doesn’t prohibit a stupid transaction people will do it in mass. This is like saying if its legal to light your cigars with $20 bills we need to start printing lots of $20 bills because we will soon see a shortage.

    Report

    • @Boonton, not my argument at all, Boonton.

      It’s that the government was engaging in a spectacularly ill-thought out program.

      It’s a program to stimulate new car sales by purchasing beat-up Mercedes and Jaguars (among other cars) then destroying them rather than putting them through Phil Long’s (or whomever’s) 120 point inspection program, fixing them up, and turning the $500 car, with a little maintenance, back into a $2500 car with only $1000 worth of parts/labor that you can then sell to some poor schmuck who is trying to get rid of his 1977 Datsun.

      Report

      • @Jaybird,

        So the car is worth $500 yet by applying ‘magic’ it could have turned into a $2500 car? Why don’t you buy this $500 car from the junk yard and make it happen? Sure the engine was destroyed but you can put a new one in for less than $500 (since, of course, the car as a whole was worth $500 its used engine has to be worth less).

        Perhaps the reason you don’t do this is that only in your imagination is auto labor actually free.

        That doesn’t explain, though, how eliminating high blue book values from the list would have made the program better.

        Report

        • @Boonton, it’s not applying “magic”, though I understand how someone who hasn’t spent a whole lot of time outside of academia might think it was.

          It’s called “maintenance”. Here’s how it works.

          You buy a beat up car. You apply your labor and some purchased parts to said beat up car.

          By doing this, you can turn the beat up car into a car that will be worth more to someone else than the price you originally paid for it *PLUS* the cost of the parts *PLUS* the cost of your labor.

          You can then sell this worked-on car at a profit!

          (Seriously. Talk to some of the used car dealers in your town when they’re not hustlin’ and you’ll find that this is how many of them got their start in the 80’s… buying cars, fixing them up, then selling them.)

          Report

          • @Jaybird, OK so do it then. The worse case is that you shell out $500 to grab an engine from a car that was junked not as part of the C4C program and you sell the car for $3000 instead of $2500. Since used car prices are ‘soaring’ anyway you should be able to pull this off. Whose stopping you?

            Report

  31. Jaybird

    Because the worst cars out there (those 1977 Datsuns that I keep complaining about) are still there.

    I thought you said the 77 Datsun got good mpg? If some are still on the road why is that a major issue?

    We’d get more benefit from getting rid of those than we would from getting rid of a number of Ford SUVs.

    How so? A Ford SUV worth $500 gets worse mpg than a 77 Datsun worth $500.

    Heck, we’d get a non-zero amount of benefit from straight up trading the Ford SUVs for those Datsuns.

    I’m not sure how. I guess the Datsun owner gets a benefit from switching to a Ford SUV but what does the Ford SUV owner get? Money? Then have the Datson owner pay the Ford owner…. Yea I suppose if we are crushing a Ford SUV the Datsun owner might as well take it and enjoy its use….but that would lower the average mpg on the road while one of our goals was to increase it.

    Look I get it. The junked cars had some value. That value was their market price. If the average junked car was worth $500 then the cost of the destroyed capital was $350m (not billion). As I pointed out, I think the true value of the destroyed capital was less than its market price (and since I believe the non-engine parts and scrap metal was still able to be sold it’s not even like the full $500 was lost). That’s a pretty small portion of the $3B (11.6% of the program) so it’s a cost I’m willing to live with in the name of removing some low mpg from the roads.

    120 point inspection program, fixing them up, and turning the $500 car, with a little maintenance, back into a $2500 car with only $1000 worth of parts/labor that….

    The problem here is that you forget why the value of the junked cars was probably on the low end rather than the high end. The value of the voucher is $4500 minus the value of the junked car. The car you describe isn’t really worth $500, its worth $1000 since adding parts and labor lets you profit by $1000 in selling rather than only $500 from selling without fixing it up. A car that’s worth $1000 is less appealing to the C4C program than a car that’s worth $500 since the benefit of the voucher is $3500 ($4500-$1000) rather than $4000 ($4500-$500).

    Now of course these numbers change if people were, in mass, turning in $5000 Dodge Rams, Mercedes and Juagars to be junked. But you have yet to explain why you consider this a possibility even worth serious consideration.

    You began this argument with $24000 per new car ($3B divided by the # of net new cars sold). We are now debating pennies on the dollar (700,000 vehicles turned in times their true economic value net whatever the scrap value of their metal and non-engine parts were)

    Report

      • @Jaybird, Your problem is that I’ve understood everything you said. If you aren’t going to get better arguments your best shot is to present the ones you have in a more confusing manner. In other words if you’re going to put lemons on your lot give them a nice wax job at least.

        Report

      • @Jaybird, In sum you are saying we could have done a bit more good by letting the used cars go back on the market rather than being destroyed so the 77 Datsun guy could buy that Ford SUV and enjoy it rather than see the SUV crushed. If there’s no value to improving mpg then this value would be about the market price of the used car that got crushed (say $500) or about 11.6% of the 3B program. The program might have been 11% better than it was. OK fine if that’s what you spent a hundred posts fighting for then take it.

        Report

Comments are closed.