There's never been much enthusiasm among mainstream economic thinkers for having the government hold down apartment rents through rent controls, an issue that voters in St. Paul may soon confront.

That's why it was interesting to see some research, on University of Southern California letterhead, that suggests that the effects of rent control aren't so settled. Bad ideas when drawn on the whiteboard might not be so bad in real life.

"Less deleterious effects than is often imagined" is the way the authors put it.

While "not as bad as you thought" seems to fall short of an endorsement, the alliance of advocates behind the push for a 3% annual cap on rent increases in St. Paul put this study on their website anyway.

But you can easily conclude from the studies they didn't put on their website that rent control isn't a great idea if the goal is to increase the supply of housing.

If government caps the price of something, the market is going to see less of what it needs. Builders are probably going to look first at other markets where the expected rate of return on their investment isn't artificially capped.

Capital can flow out of the rent-controlled markets, too, with property owners cutting the maintenance budgets, putting off capital improvements and so on to accumulate capital that they will just invest somewhere else.

Owners can also kick out the renters and convert the buildings to condominiums. In one recent study, owners of small apartment buildings got around their local rent control ordinance by moving into their own apartment buildings, forcing out the tenants.

One thing that won't be allowed in St. Paul, at least in the plain reading of the proposed ordinance, is using the turnover of an apartment to a new tenant as a chance to raise the rent by more than 3%, maybe to catch up to the broader Twin Cities market.

To their credit, the advocates behind this proposal in St. Paul came up with one provision that allows room for greater annual rent increases if there is some good reason for them, like to bring an apartment up to code or cover "unavoidable" increases in maintenance and operating costs.

The proposed ordinance establishes a regulatory regime that looks a little like that of the electric power industry. The city would decide what's a permissible expense that makes a greater-than-3% rent increase OK, so the landlord could continue to earn a "fair" or "reasonable" return.

Our state utilities regulator oversees a handful of electric power companies. No idea how this not-quite-big city would have the regulatory bandwidth to weigh in on the fairness of thousands of proposed rent increases.

This kind of messy problem is what anyone can read about by sifting through the studies on rent control. Once a city applies that kind of price regulation, a lot of odd things start to happen, including, as one economist put it, how a city just "freezes."

Left on their own, property owners renovate their buildings and build new ones, tenants look for a better place to live or maybe move to a cheaper one. With rent control, a lot of that shuts down.

You can read about some of these effects in a couple of high-profile studies of rent control, in San Francisco and in Cambridge, Mass., the Boston-area home of Harvard University and the Massachusetts Institute of Technology.

A Stanford University research team looked in San Francisco at what happened after a 1994 ballot initiative took away a rent control exemption for small apartment buildings. A lot of the units later got taken out of the rental market, through condominium conversions or building replacements. Tenants who kept their apartments did catch a break. But, as expected, rents increased when the supply contracted in the overall market.

One of the policy goals was to prevent the gentrification of San Francisco neighborhoods. Instead this policy seemed to accelerate it.

In Massachusetts some MIT economists looked at what happened in Cambridge when controls got lifted.

They found there was a flurry of renovation and investment as values popped. Rents and property values didn't just increase for the buildings that had the rental caps removed, they increased for nearby properties, too. That suggests that much of the cost of rent control was borne by property owners the law didn't even apply to.

St. Paul is a different kind of city than those two coastal tech towns, as the median household income in Cambridge is nearly twice that of St. Paul and San Francisco's is even higher. And in St. Paul the problem for a lot of renters might not be the threat of rental increases so much as having trouble making the monthly rent right now.

Most St. Paul households are renters, and most of them earn less than 60% of the Twin Cities median household income, based on research of the Minnesota Housing Partnership. Of those households, four out of 10 paid more than half their income in rent. That suggests the problem in St. Paul is one of low income and not just the cost of rent.

The authors of that USC paper acknowledged that rent control is a blunt tool that can misallocate capital in housing. Yet, cleaner solutions, like subsidizing landlords to keep down rents, look like "a politically impractical giveaway."

The one practical idea that shouldn't be all that controversial is just to do whatever it takes to build more housing.

In a case that somehow seems emblematic of how well that is going in St. Paul, there's been a particularly drawn-out permitting process for a proposed 288-unit building on Lexington Parkway near the headquarters of the Amherst H. Wilder Foundation.

It seems like a great site for a new apartment building, with both a freeway and a Green Line light rail station close by. Yet there's still no new building there, on a site that's been vacant so long I can't even remember what used to be there.