Auto buybacks backfire

In my post, Gun Buyback Misfires, I pointed out that a) gun buybacks encourage people to turn in old, low-quality guns that are unlikely to be used in any case and b) gun buybacks can encourage people to buy and hold more guns because the buyback is a form of insurance, if the gun gets old or stops working you can sell it to the police.

In an excellent post Steve Levitt points out that Alan Blinder’s proposal for auto buybacks suffers from exactly the same problems.

…the majority of vehicles that are turned in will not have been driven much, if
at all. Indeed, I suspect one of the most visible responses to this program will
be a new market for mechanics fixing up cars that don’t run at all just enough
so that they can be driven to the government’s lot to collect the cash.

The biggest problem with this policy, however, is the way it distorts long
run incentives. Let’s say the rules of the program say that a car must be at
least fifteen years old to qualify for a big government subsidy to scrap it.
This gives powerful incentives to people with twelve-year-old cars they were
planning on scrapping to keep driving them for three more years to collect the
government bounty. Instead of reducing the number of clunkers on the road, this
program could actually lead to an increase!


Both Blinder and Levitt miss the politics of this issue. in my city, Atlanta, we've got a traffic pollution problem, plus cheaper license plates for "antique" cars > 25 years old, emissions insepection exemptions for senior citizens with old cars, and emissions inspections exemptions for anyone who spents at least $750 to repair their clunker without succeeding in improving its emissions. not to mention property taxes based on car value, which penalize new cars compared to old ones.

So, across the board government policy is subsidizing 15-30 year old clunkers, while taxing and emissions-inspecting 3-4 year old cars with modern emissions controls.

My 80s Dodge Colt gets 38 mpg, actual number off the odometer that I have to know because the gas gauge stopped working almost immediately, so I fill up using miles since last filled. Not exactly a clunker. It can't put out more carbon than is in the fuel no matter how much a clunker it might be, in any case; so mpg is the only number in play.

Plus the carbon footprint improvement in not building a new car.

Not that global warming is real in any case.

plus cheaper license plates for "antique" cars > 25 years old

NJ has historic plates too for antique cars but there's limits on how many miles you can put on them per month. Most cars I see with these plates are true antique cars...not junkers people are plating for cost savings.

I think the place to target the incentive is in the registration fees. In NJ you have to register your car each year for about $42. Add a fee for older cars and turn it into a lump sum credit for registering new cars that get above average mileage. The only reason to register a car is to drive it so you won't have the issue of people fixing up junk cars unless they honestly want to use them on the roads.

The sentiment behind the proposal seems sound, and while there are clearly issues with incentives that others have pointed out, I have to ask, are there no solutions to these problems with Blinder's proposal? One of the issues that Levitt raised was people holding on to cars for a few extra years just to take advantage of the subsidy. I'll ask the same questions I asked over at the Freakonomics Blog:

Couldn’t you deal with the issue of age of the cars by establishing a minimum, or rather a maximum, MPG that cars have to have? In other words, if the guidelines say that the cars can’t get more than 25 MPG, wouldn’t that render the age of the car unimportant and thus prevent people from holding onto the cars for a few extra years? It’s not like the MPG is going to magically increase, after all.

And why would there be a constituency for the program over the long run? If the MPG guideline is instituted, over time, wouldn’t the cars under this guideline eventually fall so that nobody would qualify?

Blinder plays a bait and switch, talking about "pollution" in most of his piece and mentioning greenhouse gasses at the end. Older cars tend to have worse mpg, thus carbon emissions, but they are dramatically worse for NOx & SOx, which is the topic of the studies he repeatedly quotes. If he hadn't mentioned greenhouse gases at the end, it would be merely deceitful and not an explicit lie. Probably an editor who didn't know the code added it.

The problem with many of your suggestions is that, almost by definition, someone who uses an old, poorly-maintained car is poor.

So when you mandate that those cars be off the road,

Countries with a little firmer worldview would just implement inspections ($) to get badly maintained old cars off the road.

or try to nickel-and-dime them off the road

So, across the board government policy is subsidizing 15-30 year old clunkers, while taxing and emissions-inspecting 3-4 year old cars with modern emissions controls.

Add a fee for older cars

you are really saying that poor people shouldn't own cars.

One nice thing about this proposal is that we don't have to worry too much about some of the hypothetical problems suggested, because we have several pilot programs in place that we can examine. The California program that I am familiar with has been relatively successful, and has been held back largely because it is underfunded.

Indeed, in California's version, the model year cut-off date eliminates Levitt's second concern. Cars are bought for junking based on their model year, to eliminate older anti-pollution technologies that weren't as effective.

As to the first concern of Levitt re fixing up older cars, that has not been a problem observed here in central California. However, that may be due to the program not having enough funding to create a market beyond the existing cars already in need of junking.

This seems like a straightforward program for an economist to endorse - incentivizing the behavior you want to see. Perhaps in a larger scale of the program, you would see problems like cars being fixed up for resale, but that strikes me as suggesting that the price might be slightly too high for market clearing of the old clunkers, not that the program itself is inherently flawed.

A version of this proposal avoids some of the problems noted above: Purchase old cars that are advertised in the classified ads. At least at first, this will largely lead to removing running cars from the market. If such a buy-back grows to large scale, it runs into the problems of raising the value of old cars.

Thus, if we care about reducing pollution then taxing pollution makes more sense. If we can bring to market the almost-ready technologies that permit sensors in the road to detect car-specific pollution, pollution taxes should be feasible at reasonable transaction costs.

The best way to remove dangerous or "bad" cars from the road? Make it costly to run them....
That is, via inspections and registrations. I have a 13 year old Explorer that I love. At 55 mph it gets 23 mpg. At 70, it gets 18. It's not a huge disadvantage for me to keep it because the cost of trading in and getting a new car (low trade in value, high cost of purchase) that gets 30 mpg is MUCH more than the cost of gas and repairs on the current car (I drive it about 8,000 miles per year).

That's $700 in savings if gas is $4 and I get 18 mpg (I usually get more), per year, on getting a 30 mpg car.

Now, add in maintenance. I pay about $1,000 a year in maintenance. $1,700 in total payments or about $150 a month. A new car will run me much more per month.

So let's factor in inspection and registration. If, at inspection, I get assessed a cost for age and low mpg, say $200, and then I get assessed $200 at registration for same, my costs jump to $2,100. Suddenly, I'm getting closer to a cost valuation where a newer, better, car makes sense. Still not there, but in a year or two when my maintenance goes up to $2,000 a year or so (it will, age does those things), and the assessments also rise, then a new car would inevitably make more sense. Let's say the Explorer turns 15, my costs are $2,700 and registration and inspection take a $600 bite (rising fees on cars over 8 years old, let's say, with penalties for gas guzzlers). At almost $300 a month in costs, and I'm close to being in the market for a new car.

It's still a stupid idea...the real cost of savings and maintenance have to exceed $450 a month for the change to make sense on a $24,000 car. Yet how many $24,000 cars are there out there? I just did a search, and not many.

I found some Chevy Hybrids at under $24, the gas savings could tote up a bit more (since I do mostly local driving) to about $1,200. Still, it's not enough to make a huge difference.

Net result? Ride out your car, if you own it outright as I do, until the cost of maintenance matches the cost of a new car payment. Then trade in and buy a new one.

If the gov't wants to make a little money and "help" the car manufacturers, then the way to do it is at registration and inspection. Free for cars under 8 years that get over 25mpg. Under 8 years and under 25 mpg - $100. Over 8 and under 25, up $30 or $40 a year.

It's punitive, it's unfair, but it would do the trick.

I was thinking about making a car donation and buying a new one. But this program convinced me to keep my 13 years old car, drive it for two more years and buy a new one with a down payment gotten from government through the cash for clunkers program.

The truth is that we can't really know how people are going to react to this program. Many of those who can't afford a new car are still driving their old cars until they save up enough for a new one. This program allows them to do this a lot faster thus eliminating an old car from traffic.
Land Rover Parts

The author pointed out well that the auto buyback attempts could be a detrimental business. While working at Chevy dealers Philadelphia I've noticed that when this discussion was on topic, the cars sells have dramatically decreased.

Steve Levitt makes some precious assessments. If I drive an old car and I am prepared to improve it with a couple of BMW parts, what's stopping me from waiting an year or two to cash the Government compensation?

Comments for this post are closed