Actually, this is just part of Vid's Space. For more, go back to the full Vid's Space.
So there's this concept called induced demand. It says if you build a new road, or add capacity to an existing road, people will change their habits and drive more, using up the added capacity. I won't claim this is untrue.
I've seen it stated several times now, however, that the extent of this effect is such that all of the new capacity is used up, resulting in exactly as much congestion as there was before. It's like how if you make a water pipe bigger, it will immediately draw more water into it. Nevermind the absurdity of enlarging a pipe while it's in use, this analogy breaks because water is an incompressible fluid, while traffic often isn't.
Sure, when a road is at capacity, the traffic may as well be considered an incompressible fluid. But there's no vacuum force sucking cars into every available lane-mile of all the rural highways of America. Not even metaphorically. There is no macro-economic, group-behavior phenomenon putting upward pressure on the traffic counts of rural roads. Unless a highway is at capacity, the traffic is a compressible fluid which is perfectly happy to exist in a low-pressure state in the absence of compressive forces.
Okay, so clearly this idea of congestiostatic induced demand is framed in the context of urban transportation planning, and my rural roads counterexample is irrelevant. Yes, in cities, essentially all new capacity is immediately filled by induced demand, resulting in nearly (but not exactly; more on that below) the same amount of congestion. This is because in cities, the overall demand for vehicular mobility is greater than the capacity of the system; roads that might have more capacity than demand are filled to capacity with traffic seeking an alternate to other roads with more demand than capacity. The limiting factor on an individual's travel (and on the total vehicle miles traveled) is how much congestion people are willing to put up with.
Note how this observed macro-economic effect is driven by individual choices. Now let's examine the hypothetical situation transportation planners like to conjure, of two cities alike in every way except for their highway capacity. Let's say one city (Broadville) has 50% more lane-miles of highway than the other (Narrowtown). Some measure of "congestion" is performed, and the transportation planners will expect the congestion is exactly the same in both cities. Folks in Broadville drive 50% farther, or 50% more often, or some combination thereof, compared to their Narrowtown counterparts. 50% more capacity means 50% more traffic. But why? Why does a person in Broadville do 50% more driving than a person in Narrowtown if they have the same levels of congestion? If a Narrowtowner and a Broadvillianite were to switch places, they'd take their habits with them, and not increase or reduce their respective weekly driving, because they experience the same amount of congestion as before the move. The exchange of these two individuals alone would not significantly impact the congestion, as they are just two out of hundreds of thousands of people driving about their hypothetical surrounds. I assert this is an impossible scenario.
The congestion can't possibly be the same in both cities. Narrowtown must have slightly higher congestion than Broadville. Specifically, driving 150 miles a week in Broadville must be exactly as tolerable as driving 100 miles a week in Narrowtown, or people would change their habits until equilibrium is reached. Even at capacity, traffic is slightly compressible, so that the tiniest difference in "congestion" results in enough of a difference that people can tolerate 50% more time and distance in the "same" congestion. Even liquid water will compress a tiny amount under some applied pressure.
Now urban planners will probably try to say almost-the-same congestion is just as bad as exactly-the-same congestion. Why build Broadville, when for the cost of only slightly more congestion we can build Narrowtown and have "safer, more pedestrian-friendly" streets and save a bunch of money on asphalt?
Because Broadville is 50% more mobile. Free to drive 50% farther, Broadvillianites have access to twice as many choices of grocery stores, dentist offices, parks, museums, erotic poetry slams, appliance stores, and sports arenas as compared to Narrowtowners. Broadville may seem just as congested as Narrowtown, yet the former enjoys more options and lower costs that come from greater competition.
That's the macro-scale argument in favor of greater road capacity and mobility despite induced demand. On a small scale, we can throw out the whole claim of induced demand leading to congestion staying the same. We can because that claim is based on measuring capacity, vehicle miles traveled, and congestion over entire cities. That argument is completely irrelevant when the discussion is about whether to widen or otherwise improve capacity of a specific segment of the road network that experiences higher congestion than surrounding roads. Frequently the reason for this higher congestion is concentrated demand (such as at a suburban freeway interchange) or a local drop in capacity, for which few or no alternate routes exist. Improving capacity at these locations might result in a modest traffic increase, as cars return to the preferred route from whatever alternates drivers may have come up with, but that actually means fewer total vehicle miles traveled, because people aren't pressured to take a longer path around the congested spot.
To conclude, arguing against a road project that adds capacity, "because induced demand", is not a sound argument. On the other hand, I am very much in favor of — in addition to adding road capacity where needed — alternative transportation options such as rail and/or bikeways, and planning different land use types close together so residents have more options without getting in their cars.
And if you oppose a road project because it will "cause more development", you really need to tell your city leaders (or whatever local government is applicable) that you oppose more development. Development is not stopped by congested roads; I've seen farmland go right on becoming housing developments despite the farm roads already being choked with traffic from the developments that came before. Development is stopped by zoning boards who say no to developers. Any other tactic amounts to making your own town unlivable, and hoping prospective new residents notice before they buy.