Here are two things that sensible, right-thinking Americans want to do:
1) Increase wages for the people at the bottom of the distribution, without throwing a lot of people out of work
2) Increase the supply of housing in order to make housing cheaper for the average American
What if there were one policy that could do both of these at once? My idea is to index future minimum wage increases to local rents. There are some challenges for making this work, but it has the potential to improve both wages and housing at the same time. Let me explain why…
How fast should minimum wage go up?
As it exists now, the minimum wage goes down over time in real terms, because of inflation. For example, the federal minimum wage in 1980 was $3.10, which in today’s terms is $9.84 (36% higher than the current federal minimum wage of $7.25). So in order to keep minimum wage at a level that allows workers to live a decent life, minimum wage needs to go up over time.
But there’s also the issue of local differences. A $15 wage goes a lot farther in small-town Oklahoma than it does in New York City. If we set the small-town Oklahoma wage too high, it’ll cause problems, including unemployment; if we set the NYC wage too low, it won’t boost people out of poverty. The clear solution is to index minimum wage increases to local conditions in some way, so that you can have a higher level in NYC than in small-town Oklahoma.
But which conditions? One suggestion is to tie minimum wage increases to local productivity. The theory here is that wage is supposed to equal productivity (remember Econ 101?), and that the reason small-town Oklahoma has lower wages than NYC is that its productivity is lower. In fact, the idea that minimum wages should have kept pace with productivity is a big reason people want to raise the minimum wage to $15 now, instead of $10, since productivity has gone up by more than inflation.
But there are problems with the productivity approach. First of all, what we actually measure is average productivity, while the Econ 101 theory says wages should equal marginal productivity. Those aren’t necessarily the same, and can diverge over time — for example, if returns to scale change. In fact, marginal productivity isn’t even something we can observe! Second of all, productivity is subject to composition effects — it tends to jump up in recessions, not because recessions make the economy more productive, but because less productive workers tend to get laid off in large numbers. And third, productivity is an average, and changes in that average may not apply to low-wage workers.
Another idea is to index future minimum wage increases to inflation. This would probably be OK. But changes in average prices don’t necessarily equal changes in the prices of the things poor people buy. If stuff the poor people buy goes up in price faster than the average, then minimum wage wouldn’t keep up if it were tied to the average. In other words, if the price of big-screen TVs and luxury cars and McMansions goes down, but the price of food and rent and bus fare goes up, we don’t want to cut minimum wage. So that’s one thing to worry about.
The best idea put forward so far is to tie future minimum wage increases to median local wages. In fact, this is the solution Biden has proposed! The reasoning here is that median wages tell us about how high we can raise the minimum wage before it starts to risk significant unemployment. The general rule of thumb is that if minimum wage is 60% of the median wage (for full-time workers) or lower, it’s not going to cause much job loss. So if minimum wage is at a safe level and it rises along with median wage, it’ll remain safe.
This is a good solution, but I think we might be able to do even better. My idea is to tie local minimum wage hikes to increases in local rent. There are two reasons to do this:
Because rent is a good proxy for the cost of living for poor people, and
Because tying minimum wages to rent provides an incentive for cities to build more housing.
Rent is the biggest single expense that low-income Americans have, and they spend more of their income on rent than people with higher incomes do. Via Eli Dourado, here’s a chart showing the percent of consumption spending that goes to rent, for all the income deciles:
Of course the y-axis is truncated, so the difference isn’t huge, but you can see that people at the bottom of the distribution spend over a quarter of their income on rent, while people at the top spend maybe one-sixth. A quarter of your income is a huge amount!
Also, inflation includes luxury items and stuff that people can make do without. But no one can make do without shelter, so it seems like if we want to make sure that minimum wage keeps up with the cost of living, we should look at the cost of the most essential items, like shelter and food, instead of inflation overall.
So indexing minimum wage to the cost of shelter would be better than indexing it to inflation. But what about indexing it to median wage? Well, tying minimum wages to (market) rent would have another added benefit: It would push cities to build more housing.
Currently, many American cities have an affordability crisis — high-income people have been moving in, but powerful homeowners dominate local politics, and these homeowners tend to fight against any new housing development. Rising demand for housing with stagnant supply has caused rents to spiral.
The solution, of course, is to build more housing, in order to accommodate the inflow of people. But NIMBYs block this at the local level. The federal government would like to get involved, but because of America’s federalist system, they don’t have many policy levers for forcing states and cities to build more housing. The best federal policy that people have been able to suggest is to hand out money, e.g. through HUD, and then making it conditional on hitting housing targets. But this is a pretty weak lever.
Suppose, though, that minimum wage increases were tied to (market) rents. Businesses that pay low wages — especially restaurants — want to keep minimum wage as low as possible. If they could only do this by lowering market rents, they would become a powerful pro-housing lobby! You’d see local NIMBY homeowners show up to planning meetings, only to get shouted down by local restauranteurs!
And the beauty of this system is, even if business lobbies DID succeed in holding down minimum wages by forcing local governments to build more housing and reduce local rents, it wouldn’t matter. Because rent would be so cheap, it would still work out in poor people’s favor; their wages would be lower, but so would their rent.
Now, there are a couple ways I see that this system could go wrong. First, in places where business lobbies managed to keep minimum wages low by keeping rent really cheap, poor people could still suffer from high costs for items like food, transportation, clothing, and home heating. One solution here is to tie minimum wage to a basket of essential goods that poor people tend to spend most of their money on — so instead of just rent, tie it to rent and food and clothing and utilities. Since rent is the biggest single item in that basket, and the one whose cost is easiest for local governments to reduce with wise policies, this would still provide businesses with an incentive to lobby for more housing supply.
A thornier problem is: Which rents? Rent differs a lot by neighborhood, and if you tie minimum wage to median rent, cities might be able to game the system by reducing housing costs for the middle-class but not for the poor themselves. So you have to have some way of measuring rent for low-income people specifically. That data could be tough to gather reliably, and cities could try to game it. I see this as the biggest challenge for this policy idea.
But anyway, I believe this is an idea worth investigating. The possibility of killing two birds with one stone — tying minimum wage to the actual essential costs faced by poor people, while also giving cities a federal-level incentive to build more housing — seems too tempting to ignore.
(Update: Of course, the policy would have to be based on market rents for low-income people, so that cities couldn’t just game it with rent control. That would include the rent charged by public housing, government-subsidized housing, and housing subsidized by inclusionary zoning, but would not include rent-controlled units.)
(Update 2: Looks like some folks in Congress are actually considering something along these lines!)