A housing bill passed the House of Representatives last week by three hundred and ninety six votes to thirteen, in a chamber that has spent years unable to agree on the date. There is a reason, and it is not that housing has stopped being contentious. I will come back to it.
First, what the thing actually does, because a bill with fifty six provisions is not summarized by its vote count.
What the federal government can and cannot do about housing
This is the necessary starting point and almost every account of federal housing legislation skips it.
The binding constraint on housing supply in the United States is local. What may be built, at what density, with how much parking, after what process of review, is decided by cities and counties through zoning and permitting. A federal statute cannot change a suburban zoning code, and Congress could not give itself that power without a constitutional argument nobody wants to have.
So federal housing policy operates on the levers it does hold, and there are four. It can spend money. It can attach conditions to money it spends, which is how it reaches local decisions indirectly. It can regulate the mortgage market, which it does extensively through the agencies that guarantee and buy loans. And it can streamline the federal processes that its own programs impose.
Every provision in this bill is one of those four things. None of it rezones anything.
What is in it
Fifty six provisions across six areas: housing supply, manufactured housing, mortgage financing, rural housing, veteran housing and community banking.
The supply provisions are mostly incentives and process. Authorization of incentives to build, streamlining of approvals for infill development on land inside existing built up areas, a program to convert abandoned buildings into housing, and grants to modernize existing homes rather than build new ones.
The manufactured housing provisions matter more than they sound. A factory built home is the cheapest new housing available in America by a wide margin, and its principal obstacles are financing and local rules that exclude it. Federal financing rules are a federal question, and updating them is one of the few things Congress can do that directly changes what a low income buyer can afford.
The rest is the ordinary furniture of a housing package: rural lending, veterans’ loan benefits, and provisions easing the regulatory load on small banks, which are disproportionately the lenders in places large banks have left.
What it is not
It is not a spending bill of any scale. Authorizing a program is not the same as funding it, and a great many authorizations in American law have never received an appropriation. The money is a separate fight on a separate schedule.
It does not address rents directly. Nothing in it lowers what anybody pays next month.
And it does not touch the tax treatment of housing, which is where the largest federal subsidies actually sit and where any serious reform would have to go.
The mechanism that produced the number
Now the answer.
An earlier version of this bill contained a provision on build to rent housing: purpose built single family rental developments, which have grown rapidly and which a number of members regard as institutional investors buying up the starter home market. The provision would have required such homes to be sold within seven years of construction.
That provision was removed before this vote.
Whatever you think of it, it was the part of the bill on which people disagreed, because it took a position on who is allowed to own housing and for how long. Removing it removed the argument.
What was left is a package of authorizations, process improvements and financing updates, each of which has a constituency and none of which has an organized opponent. Manufactured housing lenders want the financing rules updated. Veterans’ organizations want the loan benefit better known. Small banks want the regulatory load reduced. Cities want money for converting derelict buildings. Nobody is against any of that.
A vote of 396 to 13 is therefore not a sign that Congress has found consensus on housing. It is a measurement of how much of a bill is left after the contested part has been taken out, and the answer is: quite a lot, and it is all real, and it is all second order.
The provision that was removed, and why it divided people
Worth setting out properly, because it will come back and because the disagreement is a real one rather than a partisan reflex.
Build to rent means single family houses constructed from the outset as rental stock, owned at scale by an institution rather than by a landlord with three properties. It has grown quickly, and the objection is that it competes for land and materials with houses that would have been sold to people who wanted to buy one, in exactly the segment where first time buyers shop.
The defense is that a house is a house. A rented dwelling houses a household, the sector has added supply that would not otherwise exist, and institutional owners maintain properties to a standard small landlords frequently do not.
The removed provision would have required such homes to be sold within seven years, which is an attempt to have both: build them as rentals, then release them into ownership. Whether that works or simply deters the construction in the first place is genuinely unsettled, and it is the sort of question where the answer depends on numbers nobody has yet.
Which is not an argument against it
Worth saying, because the cynical reading is available and is too easy.
Second order measures work. Financing rules determine what gets built at the bottom of the market. Process delays are a genuine cost that lands in the price. A program that converts derelict buildings produces actual dwellings that actual people live in.
Legislation that a large majority will vote for is also legislation that survives a change of administration, which is more than most housing policy manages.
The honest position is that this is a useful bill that will make a modest difference over years, passed at a size that reflects the removal of the thing that would have made it contentious, and that both halves of that sentence are true at once.
What to watch
Three things, and the first is the only one that determines whether any of this happens.
Appropriations. Every authorization in this bill is a permission to spend that requires a separate bill actually providing money. Watch the housing lines in the next appropriations cycle, and if they do not move, the fifty six provisions are a statement of intent.
The Senate, which has its own version and its own members with their own views about build to rent, and where the removed provision may well return.
And the manufactured housing financing rules specifically, because they are the provision most likely to change what somebody can buy, they are administered by agencies rather than appropriated by Congress, and they will be implemented quietly over the next eighteen months by people whose names will not appear anywhere.
