Streets and roads are not like teachers and doctors

Neil Salmond has been a Strong Towns fan for over two years now, and recently helped bring Chuck and Jim to Vancouver for two curbside chats and a tour. (Thanks also the City of Surrey, Translink and Simon Fraser University for making the trip possible.)

A renewable energy consultant from Edinburgh, Scotland, Neil is trained in finance and climate policy, and now lives in Vancouver with his wife and their baby son. Neil still finds it amazing that simply arranging buildings and streets differently can either create economic value (and lower emissions) or destroy fiscal (and environmental) economic bases. As he puts it, we are turning the Earth into Venus all because transport engineers think 12ft lanes and 20ft setbacks make for safe streets.

He sporadically posts to three tumblogs: Mid-Rise, Mixed-Use and Stroad to Boulevard (on buildings and streets, respectively); and the Techno-Optimist, where 3D-printed electric robotaxis do feature (but flying cars definitely don't). Follow @neil21 on twitter, or seek him out on to debate macroeconomics (Chuck does).


"It was too simple. He’s basing his math on assessed valuation alone, not sales tax or other economic considerations. That might be a proxy in Minnesota, but not here in Wyoming."
Mike Glode, who runs Shively Hardware in Saratoga

That quote is a response to a recent Strong Towns curbside chat, reported here. I’ve heard a variation of Mike’s critique a couple of times here in Vancouver, when justifying the latest provincial highway project. Here’s a paraphrased straw man, for the sake of argument:

'You can't demand a Return-On-Investment [ROI] calculation for transport infrastructure, because it's too complicated. The economics are all intertwined.

'Some of the taxpayer investment return comes directly back to the municipality as property tax, but other revenue - like improved income and sales taxes from all the businesses and workers whose profit and wages are improved by the transport investment - that tax take first flows upwards to the province and the Feds, but then comes back down as transfers from those higher governments.

'It's a public investment because it's a social investment, like teachers and doctors. The benefits are more nebulous. You can't calculate an ROI.'

I call bullshit. Streets and roads are not like teachers and doctors. You can calculate an ROI, and here’s how.


Toll them. If there’s an income benefit to people or businesses from using the road, then they’ll use it. If not, they won’t.


Now, maybe - for whatever good democratic reasons - the taxpayers and voters want to subsidize the toll for all the drivers using the road. You should still be able to calculate (and advertise!) the hypothetical fee per vehicle that you’re waiving. The fee would return your hurdle ROI over the lifecycle of the road (including all operation, maintenance and construction debt repayment costs).

Note: chances are more people will use a road that’s free at the point of use, so you might want to accelerate your depreciation, as compared to a tolled road.


Property tax. A street is a platform for value: it is an outdoor room that adds value to the surrounding properties, by providing safe and pleasant access. The proportion of annual property taxes received from surrounding properties that is available for street maintenance must at least match the annual depreciation expense on the street.




Whatever their scale, note that you can both minimize the maintenance cost of a street and improve the surrounding property values by dedicating most or all of the surface to intermittent vehicles (transit) and slower, lighter ones (feet and bicycles). Many, heavy, faster vehicles (cars) will wear the surface more quickly, and detract from property value.


If the thoroughfare under ROI analysis is neither a street or a road - if it has features of both, such as multiple or wide lanes for high speed travel, but also sidewalks, intersections and property access - then the thoroughfare is a stroad and it will offer a poor ROI.


A stroad offers neither the toll-able efficiency of an A-to-B road, nor the property-value-boosting platform of a street. Start-stop congested drivers will be frustrated, cyclists will be only the few and fearless, and pedestrians (aka transit-users) will be few and uncomfortable if not entirely absent.

Tolling intersection-pocked stroads is technically difficult - not to mention politically out of favour - so most municipalities have treated stroads as streets, with the maintenance charges falling on surrounding properties. Those are of course the very properties whose value is undermined by this pedestrian-repelling land use. When these property tax revenues are - predictably - found wanting, municipalities are left begging for transfers from higher up government, citing all the spurious and nebulous ROI benefits in the straw-man case above.

(British Columbian municipalities’ latest begging came under the banner of “fairer, more economically responsive revenue tools” to give municipalities a “greater share of the benefits of overall economic growth”, i.e. other tax revenue. The Strong Fiscal Futures report didn’t mention infrastructure ROI, zoning reform or road tolling once.)


Analyze transit ROI in a similar way to roads: what would be the full cost to move a person. If the passengers paid the full amount of ongoing O&M costs, you would have a farebox recovery ratio of 100%. However, you would have to add to this the cost to pay down the debt that purchased the buses or trains, and laid the tracks or asphalt or dug the tunnel.


Now, just as with the road, your electorate and taxpayers - within and above the municipality - might have reasons to want to subsidize transit. Simple geometry for example suggests that moving people en masse in vehicles that don’t park at passenger destinations will take up less space than moving people in individual cars, and so allow more space in your city for (tax paying) economic activity. Local economic considerations might favour subsidizing transit before roads, as less money per trip would leave the local economy for fuel and car-parts purchases. Democratic equity might be a consideration, as transit is available to both children and the old, and to other voters without a drivers license, or the means to purchase, maintain and fuel a car. Environmental considerations might also be a motivation, as mass transit obviously consumes less energy per person-trip than private vehicles.

There is sometimes confusion about whether transit should be treated as a ‘street’-like investment, with costs recovered primarily through property taxes. A rapid transit stop can bring a tonne of value to a neighborhood, which can be captured through property taxes: this is precisely how streetcar neighborhoods across North America were originally built. However this is more analogous to the value creation of the original railroads: once the station arrives, surrounding land value rises. The dominant function of the transit investment is connection, not creating a local value platform. The connection benefit of transit is felt on the entirety of the surrounding neighborhood, not on a single street: indeed, streets that aren’t directly on the transit route may see a greater uplift. This is very different from the access and value-platform function of a street.

Now, the subsidy for transit could well come (in part) from local property taxes, but it should be clear and transparent that this is the case, so that citizens can compare that value to funding for schools, libraries etc.

Teachers and doctors

I really don’t know where you’d start with ROI here. Teaching to the test, and assessing on grades, seems to be generally unpopular now. Maybe we could do some kind of a bankers-like withheld bonus scheme, paid if students are gainfully employed a decade after graduation? Teachers and doctors don’t wear out with use along a quantifiable depreciation curve in the way that streets and roads do, so to me there’s a stronger case for the social benefits of education and healthcare to be exempt from cold hard ROI calculations.

But streets and roads do wear down predictably over time, and have to be repaired. This predictable maintenance can be paid by users in the case of roads, or by the beneficiaries of access in the surrounding properties in the case of streets. Both can be subsidized, but the subsidy should be explicit and transparent so that voters can make informed decisions.

Waving your hands and claiming some magical unidentified multiplier effect will pay for all the maintenance is fiscally unsound, and bad governance.