Hacker News Reader

Comment by tonyedgecombe | Hacker News Reader

How does making cars less efficient help the rural poor?

By lowering the cost to manufacture a vehicle by reducing the artificial constraints imposed by environmental regulations, more people are able to purchase vehicles at a lower cost, leading to an overall higher standard of living for the poor who couldn't otherwise afford the aforementioned vehicle.

But then they end up spending even more in gas than they saved.

Also, poor people aren’t buying new cars, they’re buying used cars. So it’ll take a while for new more efficient cars to trickle down, and by that time they’ll have depreciated well past the theoretical delta in price.

It could also be that regulations lead to a superior and different, not necessarily more expensive design.

Good theory, but you’d have to know the incremental cost of the lower emissions on a per-vehicle basis, and what effect those costs have on purchasing decisions (taking into account alternatives and so on). First, is there any basis for assuming a more fuel efficient car is in fact more expensive to produce on a per-unit basis? For example, a smaller car will be more fuel efficient than a larger car using the exact same technology. Of course there are probably some approaches that do cost more, like using more expensive materials that are lighter in weight or something. But even so, let’s talk percentage difference. Is the MSRP 1% higher? 2%? It’s pretty preposterous to suggest that someone were talking about here (lower class rural) is going to be buying a new car anyways, and that a small increase in the purchase price will swing that decsuson. If you make minimum wage and you want a new F150, whether it’s $35k or $36k ain’t gonna matter - you can’t afford it anyways. So we’re talking used market down the road, and that higher initial purchase price will be discounted at the rate used cars depreciate, so our couple of percent increase is now a fraction thereof.

In economics, everything is on the margin. In engineering, constraints tend to define tradeoffs.

If we look at the population, then for everyone that needs/wants a car of size X, age Y, and cost Z, then fuel efficiency regulations that increase the engineered complexity of the vehicle will push up the cost and mean the consumer has to trade off with a smaller car, older car, or some other aspect.

Discretizing the population into poor people that can't afford a truck and rich people that can is meaningless when there is an entire distribution of people's incomes and spending preferences.

But the assertion was that environmental regulation was regressive, in that it disproportionally impacts those on the lower end of the socioeconomic spectrum. So I was trying to show, a bit ham-fistedly maybe, that incrementally more expensive new vehicles do not impact the lower socioeconomic rungs very much.

> First, is there any basis for assuming a more fuel efficient car is in fact more expensive to produce on a per-unit basis?

Great question! There's actually some pretty good literature on it from energy and policy economists. The auto industry, the EPA, and the NHTSA all pull from very similar general background estimates.

"A 1991 study performed by Greene[1] concluded that, for every one mile-per-gallon increase in vehicle fuel economy, the average, per-vehicle cost would be within the range of $100-200 dollars, in constant 1985 dollar terms, and corresponding to a range of $200-400 in 2010 dollars. A review of CAFE by Klier and Linn[2] found that numerous other studies have confirmed these figures. Using the range found in the literature, the projected 12.7 MPG improvement in fuel economy would cost on average somewhere between $2,500 and $5,000 per vehicle."

So... that's the industry perspective. But that's a simple review of the background literature, so it's not that radical. The EPA and NHTSA dip only slightly below the low end, getting down to $2,000 using additional assumptions on that baseline, such as learning curves from past R&D making newer developments cheaper. But the regulators still estimate a positive cost per vehicle. The debate between regulators and industry is the extent of that cost, not whether that cost exists at all.

It's a fascinating area of research, notwithstanding the minefield of biases one has to sort through.

[1] Greene, David L. (1991). “Short-Run Pricing Strategies to Increase Corporate Average Fuel Economy.” Economic Inquiry 29(1): 101–114.

[2] Klier, Thomas and Joshua Linn. (2011). “Corporate Average Fuel Economy Standards and the Market for New Vehicles.” Annual Review of Resource Economics 3(1): 445-462.

> is there any basis for assuming a more fuel efficient car is in fact more expensive to produce on a per-unit basis? For example, a smaller car will be more fuel efficient than a larger car using the exact same technology.

This comes up pretty frequently in car discussions. Making smaller cars helps you meet American fuel efficiency requirements, but it's terrible if you also want to (have to) meet American crash safety requirements.

Great question! The answer is that it won't always, but it will sometimes. Because this is very counterintuitive.

Here's my back of the napkin model, caveat emptor.

Say we're calculating total cost per mile. What inputs do we need?

Fuel price, fuel efficiency, amortized vehicle cost, maintenance/repair costs... both of those last two are dependent on length of ownership in both time and miles, and how old the vehicle was when it was purchased. Purpose of the vehicle is a secondary factor too, affecting efficiency and wear/maintenance.

I'm going to ignore or hold constant some of those but toy with others.

Say fuel is very expensive at $10 a gallon, and also there's a lot of low hanging fruit to increase the efficiency of a vehicle. Say you can get another 5 miles per gallon over your 20 mpg by raising the cost of the vehicle by only $500. And say you don't significantly increase its complexity or repair costs while doing so.

Before you drove 100 miles per week at $50/wk. Now you drive 100 miles per week at $40/wk. You plan to own the vehicle for 10 years. You pay off the difference and get pure profit by the end of the first year.

Sometimes fuel won't be $10 a gallon though. Let's say fuel is cheap, and the low hanging efficiencies have already been exploited. Let's say fuel is $2 a gallon, and you can only get wring 35 mpg up to 36 mpg for another $500.

Now how long does it take for the improved efficiency to pay off the premium?

Just over 60 years.

Increasing the miles driven per week pushes it back the other way. It all depends on your assumptions.

Basically we're looking at a benefit to the driver whenever:

    (price_per_gallon * miles_per_week / old_miles_per_gallon) - (price_per_gallon * miles_per_week / new_miles_per_gallon) >= increased_vehicle_cost / lifetime_weeks_owned

If it were true that more efficient cars were cheaper overall, people would prefer to buy them without any regulation required.

That regulation is required implies that is not the case.

Regulation being required is only a sign that some of the companies aren't willing to produce such cars without regulation.

People aren't always rational about price. You might want a more efficient car, but those are newer. You have $1000 and your older car just broke. The repairs cost more than the money you have, and you need to get to work. You would like a nice, efficient car but in your price range, you get gas guzzlers. Oh well, it is better than losing your job.

There are lots of factors like this folks face when purchasing things. Overall cost isn't always a factor even when it should be in a perfect world .

Cheap cars aren't gas guzzlers though. There are plenty of cheap cars with good fuel economy. The absolute cheapest new car you can buy in the US today (Nissan Versa) gets 31 city/39 highway. If you have $1000 you can buy a 20 year old Corolla, they get 31 city/38 highway. The main problem is not cheap cars being gas guzzlers, it's everyone insisting on buying far larger vehicles than they actually need.

One crucial factor to consider is that people are really, really bad at long-term planning - even if a more efficient vehicle would pay itself off relatively quickly, many people might go for the less expensive option, thanks to the immediate reward.

Generally speaking they don’t pay themselves off. I buy $1-$3k cars every ~4 years. No amount of gas savings would justify a $30k car. It’s simply opulence.

Reply on news.ycombinator.com