AFAIU the market for used cars and motorcycles in America nowadays operates under the “market for lemons” principle, causing such artifacts as the big drop of the price the moment you drive off the dealership yard.
Hence, question - how accurately is it possible to estimate the value of a used car or motorcycle based on the basic known statistics (year of production, mileage) and based on whatever info that could be harvested by a team of mechanics mucking around inside its internals? Let’s say we will define “value” as composed of two parameters, the number of miles the vehicle will drive until it is sent to the junkyard and the amount of money that will be spent on maintenance.
Obviously, I understand that no estimate can be completely accurate. For some areas, like the outcome of the roulette, the most accurate “estimate” would be a random guess and no amount of further research will improve on that.
Now let’s come back to the cars and motorcycles. Which of the following options best describe the state of the art on the evaluation:
the way a good mechanic will routinely evaluate a used car, e.g. when buying it, is already as good as is feasible given the randomness inherent in the future behavior of the equipment
the typical evaluation is not so good, but it could get much better if some well known procedures were followed, and these procedures are not followed because they are now too expensive
While there are several sites, publications, etc. that provide estimated “blue book” values, the reality is that the market ultimately establishes the price. The old adage that it’s worth what someone will give you for it prevails. While the aforementioned sources may provide some very general values which may be helpful, the real value is what someone will give you for it.
what does this have to do with my OP? My question is not about resale prices in the market as it exists today. It is about estimating those two parameters I have mentioned, neither of which have anything to do with blue book prices.
ETA: maybe estimating “money on maintenance” is kind of vague. It would be better to reformulate that as “expected number of maintenance procedures by type”. E.g. we estimate the car will require three engine tune-ups and one brakes overhaul in the next 7 years that we think it will remain functional.
I don’t think this is a very useful definition of “value” in the used car market these days. It’s not like the old days when you really could expect a car to go some number of miles before it’s worn out-- these days most cars can run indefinitely if properly maintained and not abused. Newer cars cost more for reasons mostly not directly related to reliability per se. Sure, we can expect newer cars to cost less to run on average, but it’s not the direct correlation it once was and is only a small part of why people pay more for newer cars.
Also, regarding mechanic’s inspections, there’s very little you can do to ascertain long-term reliability-- inspections are more about catching existing problems. A good mechanic can make some educated guesses as to how the car has been treated, and maintenance records can help, but to some extent you’re always just crossing your fingers a car hasn’t been abused by a previous owner.
GreasyJack, thank you for your response. I would like to focus on two of your claims.
My concern is that I heard an opposite claim elsewhere. I heard it claimed that cars are so made that they will serve for N miles (not necessarily without expensive maintenance, but still) and at some (indeterminate, depending on luck) point after that completely fall apart beyond repair. So are you sure about your claim here? Will anybody else agree or disagree with this?
so suppose I wanted to create a procedure to discover traces of this “abuse”. How would I go about that? Is this a matter of changes in the structure of the metal that would require X-rays to discover? Or what is the actual, material consequence of this abuse and how could we try finding it?
Where did you hear that? My first guess is from a habitual malcontent with no actual automotive knowledge or expertise who has a penchant for conspiracy theories. What you’ve related CAN happen, depending upon a number of variables, but to declare that cars are intentionally designed to last only so far (which I believe is what you’re talking about) is BS. Demand evidence for that claim (but don’t hold your breath).
Wow. This is way more complicated than it needs to be.
Let’s use the example of an engine whose life has been shortened by not enough oil changes. If you really want to know if the crankshaft bearings show significant wear, they can be inspected by removing the oil pan and the bearing caps. This alone costs hundreds of dollars, thousands on some models. I’ve never heard of trying to use X-rays for this, and I don’t see how they could yield much information given the preponderance of metal in and around the crank. Far more cost effective is looking for evidence and clues of maintenance done, or maintenance not done.
An experienced mechanic can usually get a pretty good indication of how well a car has been maintained. Maintenance records, if available, can be a great help. If it’s not possible to get a reasonable assessment, then the choice is to take a gamble and buy the car, or play it safe and don’t. If you’re looking for virtual certainty, it would cost more to achieve that than to buy the car.
The more info you have, the better a decision you can make. Still, many decisions have to be made with less info than we’d like. Buying a used car is often that type of situation.
when you say that this specific inspection procedure costs hundreds of dollars, how does the cost break down? Is it a just a matter of it taking lots of billable hours of work by the mechanic? Or is it a somewhat destructive procedure that requires replacement of the expensive parts that it damages?
ETA: so far, my impression from Gary T’s explanation is that the answer to my OP is “option 2”. I.e. yes indeed there are a bunch of procedures that we could conceivably use to get better estimates than what we do now, but nowadays we just don’t know how to do these procedures cheaply enough. Nevertheless, maybe subsequent explanations in this thread will prove this conclusion invalid.
Why would you be wiling to pay for that level of mechanical investigation? Bottom line is that a car is going to sell for some amount in the neighborhood of the book value for that make/model in a particular part of the country. That’s how dealers price their cars, that’s how dealers determine what they will offer for your trade-in, that’s how private sellers determine their asking price in the real world–at least for vehicles for which there is a book value. You could spend thousands of dollars for a mechanic to take the vehicle apart bolt by bolt, but other than peace of mind, you get little for the money. The mechanic’s opinion that the car is in excellent shape or is crap isn’t going to affect the sale price by enough to cover the mechanic’s labor. In some cases, the cost of that level of mechanical inspection may be more than the value of the car.
I’m not sure what your goal is here. Your definition of “value” relative to how cars and motorcycles are valued in the real world is so bizarre there is really no way to properly answer your question as it based on a wildly erroneous premise. You are asking (it appears) if there is some level of extraordinarily detailed mechanical inspection that will allow you to very precisely and accurately peg the re-sale value of a used vehicle.
The answer is “not really” for all the reasons the people responding to your question have indicated, including SanDiegoTim’s answer which speaks to issues of supply and demand varying across markets. An experienced vehicle re-seller can usually establish a fairly accurate projected value range after a general inspection of the vehicle, but once that vehicle goes on the sales lot it can sell for well under or over his projections based on the current market demand for that product. The time, effort and cost it would take to do a NASA level mechanical inspection on each vehicle is not practical or economically feasible, and even if it were done there might still be undiagnosed failures.
You are looking for a level of certainty that does not exist in the real world.
Oakminster and astro, I am interested in the underlying reality of the matter and not in the current market valuation of this reality. Maybe the current market is just way off.
Recall that Michael Milken got his billions and his jail sentence by improving the way his industry did valuation of certain assets. He didn’t just sit there and trade according to the traditional inaccurate valuations. Instead, he came up with a way to do valuations better, closer to the underlying reality.
If the accurate car evaluation procedures are feasible but too expensive now, conceivably they could be improved and made less expensive. Then, if the validity of these hypothetical less expensive advanced procedures were to become common knowledge, prices of the used cars would shift away from the market for lemons and towards fair price for the real expected utility of the car.
For particular makes and models there are sometimes common problems that occur around a certain mileage. These days the mileage is pretty high. Over 100,000 miles in a car you can figure at a minimum some major accessory like the alternator or starter may need to be replaced before too long. A mechanic can point out obvious issues like rust or oil leaks. But the really expensive things like engines and transmissions are a just a gamble. When you get to a mileage where those parts are likely to go, the car usually isn’t worth that much anyway.
For motorcycles, there isn’t much to go on at all. The sale price of used motorcycles has more to do with demand than operating and maintenance costs.
I see what your’e getting at there. There are resales now that include warranties for cars that have been certified by the manufacturer. Some degree of inspection to ensure the quality of the car is provided, and the rest of the appeal comes from the warranty. This is done with those cars for which this is economically viable. But for many cars beyond a certain mileage the cost of warranty is prohibitive, and that’s where the value comes from. Anything that doesn’t fit in that category is sold for considerably less, because there is no way warranty them for a reasonable price. No amount of inspection will reveal certain potential failures. I think car companies consider these factors more than they used to. They are the ones who are in the best position to estimate the lifespan and costs of their cars, and profit from it.
The thing about this, though, is that modern cars are getting to the point that cars are to a certain extent warrantable indefinitely. See for example Chrysler’s lifetime drivetrain warranty (or the other extremely long drivetrain warranties), where they’ve basically realized if you can hold someone to the condition of keeping up with maintenance, warranty claims on the engine and transmission are essentially nil after the first few 10,000’s of miles. And some of the extended warranty companies will warranty damn near anything made in the last 20 years, and it’s still a rip-off!
The fundamental fallacy the OP seems to have is that there’s any relation whatsoever between the use value and the exchange value of a used car. The difference between a $500 used car and a $4000 used car might reflect a significant difference in reliability, but not usually between a $4000 one and a $10,000 one. Above the bottom-of-the-barrel range, newer cars are more valuable simply by virtue of their newness. This may reflect less wear and tear (like shinier paint, or cleaner seats) or more up-to-date styling and technology, but not even necessarily. The used car market is not always a rational system, but then if car buyers were always rational why on earth would anyone buy a new car when they could buy an equally functional used one for thousands less?
But at the same time he *may *have a point. In effect he asserts that all used cars are retail priced as if they were lemons. IOW …
Even if they’re at a high absolute price (e.g. $10K), they’re priced as if, for a car of that age and model (e.g. 4 yo Camry), that particular vehicle will have major problems soon and the buyer is wary of that. And so demands a discount from the “right” price as insurance against the unknowns.
The very existence of used car warrantees, and their continuing popularity despite well publicized horror stories about them being rip-offs, indicates a lot of people do beleive that buying a used car has a large element of risk. The OP’s idea that IF we could reliably identify the specific vehicles which lack that risk, THEN we could sell those at a premium to their riskier peers, makes complete sense.
A pal of mine is in the rent-a-scruffy-car biz. And years ago I used to be in the car hop-up/repair biz. As such I’ve been to & participated in many auto auctions. The practical reality today is cars are bought and sold on the wholesale market just like so much gravel. Very little of the wholesale price is determined by the specifics of the vehicle within the price appropriate to the model, year, and mileage. IOW, 4yo Camry’s go for $7000/ton, period. No wholesale buyer is going to pay more ever and the only time they’re gonna pay less is if by chance nobody else at that auction that day needs any 4 yo Camry’s and they bid low & get them cheap.
Where the OP goes totally wrong is in assuming that pricing them like gravel is wrong. Mechanical reliability is a statistical property of the fleet. It isn’t a predictable property of the individual vehicle. At least it isn’t on the good side. Identifying a twice-totalled once-flooded beater isn’t hard. Identifying the one owned and driven by the little old lady who *didn’t *let her teenage grandson drag race it on Friday nights is not only infeasible today, it’s (IMO) impossible even with a magic CAT scanner that could see & analyse everything.
The OP is right there’s a market failure here. Retail buyers are more afraid of buying into hidden catastrophic expenses than they rationally should be. But the solution is something more like insurance, not a magical process to identify the risk-free vehicle needle in the used car fleet haystack.
The used car market can be really irrational at times. Checking local cars on Cars.com, as an example, a used Toyota Matrix will usually sell for at least a thousand more dollars than a Pontiac Vibe of the same year and mileage, even though they are the same car, and were built in the same factory in California, with only cosmetic differences. Another example would be that AWD Ford Fusions, and AWD Lincoln MKZs, of the same year and mileage, usually are priced about the same. Both cars mostly the same underneath, but the Lincoln has bigger engine, and much nicer interior.
I am unclear here as to why LSLGuy considers these warranties to be a rip-off. If they are all overpriced, why doesn’t the competition (perhaps started by LSLGuy if he were to manage raise funds) push the prices down? Or are the warranties high priced because of the threat of the undetected lemons? Or because of the threat of catastrophic accidents unrelated to quality of the car, like running into the pool and then lying to the insurance examiner?
I was about to say something like that, then didn’t know if I was just repeating rants about warranty rip-offs. I think it’s all more oriented toward profit for the warranter than any real value added to a car.
I think it’s clear that the perceived value of cars is based on supply and demand, and the demand just isn’t based on cost per mile. The hard part for **code_grey **to figure out is not how to determine the cost per mile, but how to turn the market into something that considers that a factor.
I’m going back to the point I made in my original post which is that all these arguments about this hypothetical “value add” for hyper inspected cars are premised on an utterly unrealistic notion of the technological and economic feasibility of making a very accurate and finely parsed distinction between “good, better, best” that would be reflected in pricing.
The cost of these micro-inspections would be exorbitant, and far in excess of any “value add” beyond what current inspections yield. Beyond this the ability of the micro-inspection to accurately pin the projected statistical future reliability for the vehicle over time is open to question.
If the item to be inspected was incredibly expensive (ie space shuttle or jet fighter or battleship etc.) and were being bought and sold in a rational market were there might be some merit to this approach, but if the cost of the super detailed inspection is more than, or a significant fraction of, the average retail value of the vehicle, and the results are still somewhat statistically indeterminate the rationale for doing this falls to pieces.
Not sure if this is what the OP meant but, I would think majority of used cars for sale that are less than a year old are very likely to be a repo (and therefore be very likely to have been abused) or a lemon that someone is trying to dump. In either of these situations the car would be worth much less (to me anyway) than a one year old car that was well maintained and not a lemon. Does book value reflect this and shouldn’t people whose less than one year old cars get totaled complain about the book value?
I totally disagree that a used car is ‘equally functional as a new car’. While the engine and drive train may be extremely reliable for a long time even a very reliable car has lots of parts that wear out. Most repairs have a significant cost not to mention time without a car and your time to get the car to and from a mechanic.