Week 3 — The Information Gap

A driver looking for parking in a downtown corridor at 20 mph travels about 30 feet per second. Three numbers govern what happens next.

Sight distance to an open space. Parallel-parked cars block the view of any space behind them. In real urban conditions — bus traffic, parked SUVs, foliage, signage clutter — the driver typically does not confirm an actual open space (as opposed to a hopeful gap) until they are 50 to 80 feet away.

Parallel-parking commit distance. To pull into a parallel space without hard braking, the driver needs to slow from cruise speed to about 5 mph and be alongside the car in front of the empty space. From 20 mph at a comfortable 0.2g deceleration, that takes about four seconds and 75 feet of running distance.

Sign placement. Almost no American city posts a regulatory sign at every parking space. The standard practice is one sign per regulatory zone, covering four to eight spaces, posted at the corner or zone boundary. By the time the driver spots an open space mid-block, the sign that regulates it is in the rear-view mirror.

Stack these three numbers and the realistic situation is unambiguous: the moment a driver can confirm an open space, the information that regulates it is no longer available to them. They have two to three seconds before the commit point and zero data to act on during the window. Their only options are to take the space and figure out the rules afterward, or to pass it up and gamble on the next block.

This is the argument the standard parking-economics framework is missing. The classical model treats the driver as a rational economic actor with full information. The empirical situation is closer to a driver acting on heuristics — there is a meter, so it must be a parking space; there is a painted curb in a color I do not entirely remember the meaning of; there are other cars parked here, so it must be legal — and discovering the actual rule afterward.

The violations this geometry produces are not driver negligence. They are the predictable output of an information environment that does not support real-time decision making. The driver who would have parked elsewhere if the time-band restriction had been visible at the moment of decision could not see it at the moment of decision. They get a ticket they consider unfair, because relative to what was visible to them at the curb, it functionally was unfair.

We call this category manufactured violations — violations the curb itself produces by failing to disclose its rules in the decision window. They are conceptually distinct from cruising for parking, which is the externality Shoup focused on. Cruising is the driver searching too long for a price-acceptable space. Manufactured violations are the driver committing to a space without enough information to know whether the commitment is rule-compliant. Both are deadweight losses to the city. Both compound. Neither is fully addressed by demand-responsive pricing alone.

The downstream consequences of manufactured violations are, in our experience analyzing midsize-downtown parking systems, larger than most cities have considered. The chain runs roughly as follows. A meaningful fraction of parking commitments produce technical violations the driver did not intend. The city’s enforcement system catches a small, essentially random fraction of those violations, depending on which officer was on which beat that hour. The drivers who get tickets perceive enforcement as arbitrary, because relative to what was visible at the curb, it functionally is. Public trust erodes. Word spreads — downtown parking is a lottery — and some would-be shoppers preemptively avoid the district. Voluntary compliance falls further as drivers conclude the rules do not really apply unless they happen to be unlucky. Spaces are held longer than the policy intended. Turnover falls. Commerce falls. Sales-tax receipts fall.

The city sees the ticket revenue on its budget line. It does not see the sales-tax loss on its budget line, because that loss is distributed across hundreds of merchants and thousands of transactions and shows up only in aggregate, well after the cause has been forgotten. The accounting that would tie the two together is rarely done. When it is done, the result is uncomfortable: the ticket revenue is a small fraction of the commerce loss the system is producing. The city is taxing a portion of the violations it manufactured and losing much larger amounts of commerce in the process.

The argument for closing this information gap is therefore not just a fairness argument, although fairness is the cleanest framing. It is an economic argument. Fixing the information layer at the curb does not just produce a better experience for the motorist — it produces a different category of revenue stability for the city. Demand-responsive pricing can do its work because the prices are visible. Time-limit zoning can do its work because the limits are visible. Enforcement loses its arbitrary quality because the rules were unambiguous at the moment of commitment.

This is the layer Shoup’s framework leaves off. It is not in conflict with what he wrote. It is the layer the technology of his era could not deliver, and which the technology of our era can.

Next week: The Curb Productivity Scale — why pricing-only reform stalls at Type II.