For as long as we have had cars in cities, we have needed to find places to park them.
In inner city areas where land is highly valued and there is a strong need for high amenity spaces, allocating land for storing private vehicles can be a highly inefficient use of land. Conversely, given the depth of car ownership and use in our cities, the need to cater for car parking in urban areas is a reality that cannot be ignored. How then should we manage the allocation of valuable urban land for car parking in a way that provides sufficient parking in an area, while reducing demand overall and discouraging use of single occupancy vehicle modes?
Part of the answer can be found in paid parking, and the economics behind it.
The rationale for paid parking
There are two key reasons for councils to implement paid parking in inner city municipalities:
1. To achieve an optimum level of parking occupancy and turnover
This ensures adequate access to parking for short-term visitors in appropriate locations (with consequent benefits for traders), while promoting efficient use of land, active travel and public transport use.
2. To address parking behaviour of central city workers
This can mean using paid parking to discourage workers from parking in inner city municipalities and then travelling into the central city, or to ensure that a fair contribution is made for the utilisation of parking infrastructure and the introduction of additional congestion in host municipalities.
A related benefit of paid parking is the revenues raised by fees, which councils can use to fund other projects and services, including parking infrastructure, roads, public realm upgrades and sustainable transport initiatives. From an enforcement perspective, paid parking typically results in better adherence to parking restrictions than time-based restrictions alone.
In some cases, paid parking can also offer a more practical solution than time-based restrictions. An example of this may be where a parking space situated outside a café does not yield the optimal turnover, but imposing stronger time restrictions – e.g. from 2-hour to 1-hour parking – may not allow patrons enough time at the café. Introducing parking fees in this instance would increase turnover while avoiding undermining the use value of the parking space for café visitors.
What, then, is the optimal turnover for a parking space? Is the ideal occupancy rate for a space 100 per cent, such that the moment a space is vacated another car takes its place?
Research suggests the optimal occupancy rate is somewhat lower.
Academic Donald Shoup argues that the ideal occupancy rate for on-street parking is 85 per cent.  This means that for every 20 parking spaces, 3 will be vacant at any one time – or roughly 1 in every 8 spaces.
He asserts that this proportion represents an efficient use of a parking resource and minimises the externalities associated with ‘cruising’ for parking. These externalities can include increased traffic congestion, vehicle emissions, and reduced safety. This concept is illustrated in the figure below.
Figure: On-street (curb parking prices) and cruising
Source: Shoup, 2007
The object of parking regulation should not necessarily be to ensure that an available parking space can be found directly adjacent to a visitor’s destination location, but rather within a reasonable distance from it, within an economical length of time. An optimal occupancy rate of 75-85 per cent is often used in inner city municipalities.
A best practice example: San Francisco
San Francisco’s SFpark is widely recognised as a best practice example of paid parking design and operation.  A response to high congestion, efficiency, safety and pollution costs, SFPark aims to make at least one parking space available per block (or 85 per cent occupancy) so that drivers can find a space near a specific destination, with minimal searching.
The program is based on information from in-ground parking censors which detect and report whether a parking space is available in real-time. Drivers are able to check parking fees and availability online, through the SFpark smartphone app, or via text message, before undertaking their journey. This gives users real time information to compare the costs and convenience (or inconvenience) of parking, against other modes of transport, such as public transport, walking or cycling.
A key feature of SFpark is its flexible pricing model. Based on demand, parking rates can be adjusted every few weeks by small increments. In high demand areas, rates are increased on a monthly basis until at least one space is available most of the time. The converse occurs in low demand areas, or until rates reach the minimum pricing threshold.
While this method has resulted in very high parking rates in some areas of the city, SFPark has been able to achieve its congestion-reducing objectives while decreasing average parking rates across the city.
The future of parking in our cities
While installing an extensive network of in-ground censors like that deployed by SFPark can be costly, the use of censor technology is increasing in Australian cities. These have the capacity to provide real-time data and updates, for use not only by drivers, but also regulators.
Integration with in-vehicle satellite navigation systems and voice command technology could see a greater level of en route decision making, which could raise further questions around regulating for safety and reduced congestion. Like it will for many other aspects of transport and city living, the anticipated arrival of driverless cars in our cities will also have implications for parking and congestion.
 Shoup, D. (2007) Cruising for Parking. Transport Policy 13 (6), 479-486.
 Visit http://sfpark.org/ for more information.