Last year I wrote a letter to President Trump outlining four principles for prioritizing federal spending in a new infrastructure bill. Last week, a preliminary infrastructure spending plan was leaked, the way things like this often do in our nation’s capital. Many have asked me to comment on the leaked document and so I’ll give you my immediate qactions: There’s lots to like, and it could have been much worse.

My nightmare scenario was another Obama-style stimulus bill ($831 billion total with $105 billion spent on infrastructure) only focused on handing out federal dollars primarily for new highways, interchanges, frontage roads and other build-it-and-they-will-come kinds of investments. This is nowhere near that. Breathe a sigh of relief.

My ideal scenario would have been something that focused on the four principles I put forth last year: (1) prioritize maintenance, (2) prioritize small projects, (3) spend more below ground than above and (4) prioritize neighborhoods more than 75 years old. There’s a little bit of that in this bill.

Spending Priorities

Half of the undisclosed amount of money (widely believed to be in the $200 billion range) would go into something called the Infrastructure Incentives Initiative. This has all the hallmarks of the worst of federal infrastructure spending: anything infrastructure-related is eligible, any government or public authority can apply, scoring is heavily weighted to induce local governments to take on lots of debt and there is only faint concern for long term maintenance costs or return on investment. Yuck!

But the plan has one provision that changes all of this: Grant awards can’t exceed 20 percent of total project costs. Wow! I’ve been on projects where the federal government paid 95 percent — an approach ripe with all the worst kinds of perverse incentives — but that won’t happen here. For a state or local government to get the federal money, they will need to have some serious skin in the game to the tune of 80 percent of the funding. If that provision makes it through Congress (count me doubtful), it would be transformative.

With state and local governments picking up 80 percent of the tab, I suspect projects will naturally gravitate towards those of the smaller maintenance variety, particularly projects that have a positive return on investment (small, underground, and in older neighborhoods). It’s harder to convince yourself that a negative-returning expansion project makes sense when you are spending your own money (and robbing from your already insolvent maintenance budget to do so). This dramatically reduces the worst incentives associated with federal infrastructure spending.

Of course, this shift has already brought some outrage, so we’ll see how it ultimately works out. Will Congress give up that much power? Again, I’m doubtful. Governing Magazine has suggested the proposal “puts the onus on states” (it does) and Politico compared it to the Hunger Games with cities and states fighting to the death over meager federal scraps:

Instead of the grand, New Deal-style public works program that Trump’s eye-popping price tag implies, Democratic lawmakers and mayors fear the plan would set up a vicious, zero-sum scramble for a relatively meager amount of federal cash — while forcing cities and states to scrounge up more of their own money, bringing a surge of privately financed toll roads, and shredding regulations in the name of building projects faster.

To say I disagree with Politico here is an understatement. (So does Strong Towns fan and CNBC commentator, Jake Novak, in his analysis.) Cities and states can — and absolutely must — make their infrastructure investments generate a positive real return on investment. The overwhelming maintenance liability they have now is because, for decades, with the federal government picking up almost all the bill, they never had to worry about whether a project made any financial sense. Now they do; with their own money in the game, there is no other option.

This shift should have happened in the late 1960’s when we had finished the bulk of the interstate system. Our delay makes it more painful, but no less necessary.

Have a Few Billion, Elon Musk

The plan also puts 10 percent of the appropriation into what is called a Transformative Projects Program. Sigh.

In practice, this would mean roughly $20 billion for demonstration projects, project planning and capital construction of transformative projects, which are “exploratory and ground-breaking ideas that have more risk than standard infrastructure projects but offer a larger reward profile.”

More risk? Since nearly all public infrastructure projects lose money — i.e. have a negative return on investment — I’m not sure what exactly is meant here by “risk,” but let’s just play along. I’m also not sure what a “larger reward profile” is, but I’d be less skeptical if the government had a chance, as an investor, to directly share in the financial upside of a successful project.

The Transformative Projects Program requires only a 20 percent match for capital construction meaning the “partnership” will be an 80 percent federal contribution in exchange for no equity in the potential success of the endeavor. Sadly, this sounds a lot like the public/private partnerships we’ve grown used to. In other words, it’s not really a partnership at all, just government subsidizing risk taking. Maybe you like that, but let’s not pretend it’s a true partnership.

The Rural Share

The proposal would spend 25 percent of the appropriation ($50 billion) on rural areas. Most of this money ($40 billion) would be given directly to governors to spend — literally provided to the governor to dole out — based on a formula that, perversely, rewards the total number of lane miles.

My home state of Minnesota will do well with this. Our former congressman — the late James Oberstar — was, for many years, the head of the transportation committee and, as a result, we have lots of lane miles (many of them rotting in the snow with no money to fix them). The notion of rewarding those who have been prolific in wasting prior funds is a time-tested rural patronage strategy. It’s unfortunate to see it in this bill.

And it’s hard to see how this money will be spent well. Rural areas are the worst infrastructure offenders, making the least productive investments out of desperation cheered on by eager cadres of consultants and grant coordinators. That being said, there are some states where we know that governors will take a different approach with the flexibility they are being given here. We’ll point out and cheer on those stories when they happen in the hopes they become a model for everyone.

New Financial Principles for a New Era

Critics are sure to suggest — as Governing and Politico have — that this bill is the federal government turning their backs on cities and states. This would be a more credible argument without some of the reforms being proposed as part of the bill. These changes give local governments more options for raising revenue and more motivation to find other responses to demand beyond simply begging the federal government for more money to build more stuff they can’t afford to maintain.

Perhaps the most radical change is a provision to give states the flexibility to toll interstates so long as the revenue is reinvested in infrastructure. It’s a welcome and long overdue baby step towards some actual pricing feedback. We can see this possibly leading to a broader shift away from gas tax, general funds and debt to actual user charges as the primary mechanism for funding highways. Standing ovation for a real, substantial reform.

I’m also elated by a requirement — this is not flexibility but a requirement — for value capture financing to be used to fund major transit projects. In other words, the next time California wants to use billions of federal dollars to make a few people who own property around new transit stops filthy rich—only to have those same people become NIMBY opposers of any development that would make the government investment actually pay off—the state won’t need to resort to local zoning tyranny to avert multiple crises.

When the federal government invests billions of dollars in a transit improvement, it must make the land around the stops more valuable, and then that value must be captured to pay for the project. For those of you who suggest we should be more like Japan, China or Europe in developing transit, there’s your main difference; they largely do that kind of value capture and we don’t. This is the only way successful transit in this country will be built at scale.

In summary, this isn’t how I would write a bill, but I’ll take it. There is enough reform here, and the proper shifting of costs and incentives, to where this proposal won’t be a disaster.

Charles Marohn, the founder and president of Strong Towns, has spoken in hundreds of cities and towns across North America. He was recently named one of the Ten Most Influential Urbanists of all time by Planetizen. In October 2017, The American Conservative cosponsored his “Curbside Chat” in Washington, DC.

This article originally appeared on Strong Towns and is republished with permission.