Would we suffer an "affordable car crisis" if significant barriers limited how many expensive new cars could be built?

Such a thought experiment seems reasonably uncomplicated when it comes to the automotive marketplace. Many of us have seen that straightforward market's workings up close, as buyers, as sellers and as traders of both new and used cars.

If regulations, or scarcity of materials, or other issues suddenly made it impossible for carmakers to produce a new sedan or SUV for every buyer who wanted one and could afford one — what would happen to the supply of lower-cost used cars (some 40 million of which are sold in the U.S. every year)?

Let's see: If there weren't enough new vehicles for all willing-and-able buyers, manufacturers would notice that they could not fill every order; dealers would see that buyers had become more eager to close a deal once they finally found the model they coveted. New car prices would rise, don't you suppose?

The folks who most needed or wanted new wheels would pay the higher prices. But other potential buyers would hit the brakes and decide to drive their existing vehicles a bit longer.

And the result of that would be fewer late-model used cars being traded in or otherwise put up for sale.

Next, folks most anxious to buy a low-mileage used vehicle would in turn bid up those prices — inspiring thriftier motorists to keep their aging cars awhile longer.

And in this fashion the automobile shortage would be passed down from one market segment to the next, like a chain reaction pileup on an icy highway — until eventually even the rust-bucket market was in crisis.

Or at least until used cars of every vintage had become less "affordable" than before.

Anyway, that's the theory. In the real world there actually are barriers to completely free production of new cars — especially regulations and requirements that make new cars costlier than they might otherwise be.

But the car market is far more uninhibited than the housing market, where the "affordability" crisis has become a preoccupation locally, nationally and even globally.

The foundation of housing trouble in our modern, crowded world is land. Namely, the impossibility of manufacturing more of it. Well-located building sites are unalterably scarce, so increasing housing density is the only way to expand housing in the heart of thriving, job-rich metropolitan areas. But in more places than not, government zoning restrictions limit increases to density (to say nothing of fees and codes and regulatory costs that drive up housing prices).

And one of the obstacles to building political will to relax those limits, cut those costs and let the private market build is that the "chain reaction" in housing is harder to see.

It's seems implausible to a good many people that building more upscale apartments, say, in stylish parts of town could somehow produce more "affordable housing." Those new "market-rate" housing units are anything but affordable, we're told, to people of modest means.

That doubts are widespread about the connection between high-rent and low-rent housing was made vividly clear in a recent commentary on these pages from Kelly Doran, a long-time, big-time Minnesota real estate developer who knows housing economics, both theory and practice. Doran criticized several new Minneapolis policies, including permitting triplexes across the city through the "elimination of single-family zoning."

Doran detailed the realistic costs involved in producing new triplex housing units, calculated that resulting apartments would typically have to rent for $2,400 a month or more, and concluded that the plan "will not work economically for creating affordable housing."

We can safely assume Doran is familiar with the notion that a "filtering" process in the housing market turns costly new housing into affordable older housing — over time.

But is that depreciation process so slow as to be virtually irrelevant?

No, says economist Evan Mast of the W.E. Upjohn Institute, in a fascinating paper published last summer and much discussed in urban policy circles. One urbanist admirer hailed Mast's work as "a glimmer of empiricism that helps pierce an increasingly dense fog of ideological theory."

Basically, what Mast did was reverse-engineer the chain reaction in housing. He tracked buyers and renters and counted up how long it takes for what is called "chain migration" to, in effect, transform expensive new housing units into vacancies in far-lower-cost units.

Mast says he identified "residents of new multifamily buildings in large cities, their previous addresses, the current residents of those addresses, and so on" — tracing long sequences of chain-reaction relocations that can be followed deep into the lower-rent districts of communities whenever a new unit is added at the top of the market.

Each move in the chain empties a housing unit, which is exactly like building a housing unit, Mast says.

"When a household leaves a neighborhood [it] reduces [housing] demand by one," he writes, "while building a unit increases supply by one — either way, the result is a newly vacant unit."

Some new and newly vacant units, of course, will be filled by newcomers to a city or by newly formed households (kids vacating Mom's basement, etc.). But "new market-rate housing construction can improve the market for housing in low and middle-income neighborhoods," Mast declares, "even in the short run."

Within five years or less, his calculations show, "100 new market-rate units create 70 equivalent units in neighborhoods with household incomes below the metro area median, and 39 [of them] in neighborhoods with household incomes from the bottom fifth."

Mast says these findings suggest that "chain migration" triggered by "a new market-rate unit" delivers "a significantly bigger effect on below-median income housing than … inclusionary zoning requirements — generally at least three times as large."

"Inclusionary zoning" — another Minneapolis policy Doran criticized — requires developers to subsidize a certain percentage of "affordable" units in every new project. Doran says it will discourage new construction. If so, Mast implies, it will prevent the emptying of more affordable units than it will create.

Importantly, Mast acknowledges that chain migration is unlikely to create vacancies that are affordable to those with the very lowest incomes. Some households simply can't afford any rent that covers the cost of decent shelter. Subsidies are likely to be required to meet their needs.

D.J. Tice is at Doug.Tice@startribune.com.