Toronto has done a study of preferences and spending by shoppers in one of its neighborhoods, and the results show that merchants have an inaccurate picture of the demand for driving and parking.

A summary of the findings:

• 72% of the visitors to the Study Area usually arrive by active transportation (by bicycle or walking). Only 4% report that driving is their usual mode of transportation.
• Merchants overestimated the number of their customers who arrived by car. 42% of merchants estimated that more than 25% of their customers usually arrived by car.
• Visitors who reported using active transportation to visit the Study Area visited more often and spent more money per month than those who usually drove or relied on public transit.
• Visitors to the Study Area were much more likely to prefer a bike lane or widened sidewalks over no change, even if this resulted in the loss of on-street parking.
• Merchants prefer the current layout of Queen Street more than a configuration where on-street parking is reduced to accommodate expanded sidewalks or a bike lane.
• A majority of visitors (53%) and merchants (64%) stated that there was not enough bicycle parking within the Study Area.
• Merchants were more likely than visitors to perceive the amount of car parking as inadequate: 52% of merchants stated there was not enough car parking in comparison with 19% of visitors.

This area must already be pretty safe and accessible by foot and bicycle. In most U.S. cities, I doubt we would find results like these. But if not, the lack of shopping by bicyclists could easily be a self-fulfilling prophecy if there are no bike lanes to begin with. The other critical factor, I am pretty sure, is whether people actually live in or near the shopping district. There are examples of U.S. cities that tried downtown pedestrian concourses that ultimately failed, but in the case of Philadelphia at least I am pretty sure they failed because nobody lived nearby.