How much does cleanliness cost in your city?

← Back to Clean City Lab

A "top-down" approach starts from the city's exhaustive accounts and identifies the cleanliness perimeter of the city. A "bottom-up" approach notes the cost of resources committed in the field.

In this new article, we will see that, although different, the two methods nevertheless lead to a convergent result. Annual per capita cleaning costs can vary by a ratio of one to four between different cities. There is therefore a significant source of potential for optimising road works, on condition that the costs can be objectified.

To assess them, the most precise data can be found in public accounts (the "top-down" approach). Identifying the costs of cleanliness by this method, as in the case of waste management, would provide comprehensive information without additional effort. We have proposed a standardisation method, compatible with harmonisation efforts underway in Switzerland and Europe, which would allow cities to be compared with each other.

In cases where specific information on cleaning costs is not automatically available through the city's accounting system, the teams in charge of cleaning develop their own tools. The costs of cleaning, maintaining the bins and collecting bulky waste are calculated by multiplying the resources incurred by their hourly costs (the "bottom-up" approach). This is also the approach used by the AVPU in France and OCU in Spain to compare cities with each other.

The bottom-up approach: an estimate of the committed resources

The Association of Cities for Urban Cleanliness (AVPU) brings together communities, in France and also in Europe, around the theme of urban cleanliness. Every two years since 2015, it has established an urban cleanliness benchmark, based on a questionnaire completed by road officials. The latest benchmark for 2019 is based on the responses provided by 35 cities (14 with less than 50,000 inhabitants, 13 between 50,000 and 150,000 and 8 with more than 150,000 inhabitants). Regarding cost, the respondents enter a single value corresponding to the overall amount for all urban cleaning activities. On average, cleaning costs are €56 per capita per year. For all cities, they vary from €23 to €100 per capita per year. The 2017 study also found that 82% of costs were attributable to payroll, 11% to operating costs and 7% to depreciation of equipment. The main cost is personnel. Part of the variability in 2019 can therefore be explained by this factor, since the number of agents hired per 100,000 inhabitants varies from 500 to 1,800 depending on the city.

The 2017 survey was carried out in parallel by the association of VKU municipalities with German cities, which resulted in an average of €27 per capita per year. These figures should nevertheless be taken with caution, because the results show enormous disparities in the number of agents per capita (there is a factor of 26 between the extremes), with an average value three times lower than in France (1 agent for 3,363 inhabitants on average in Germany compared to 1,003 in France).

In Spain, the consumers' association OCU also makes comparisons every four years on the cleanliness of cities and the costs of cleaning. Costs vary from €32 to €106 per capita per year between cities, with an average of €54. OCU simultaneously surveys residents on their perception of their city's level of cleanliness. Interestingly, based on this data, the level of cleanliness does not correlate with costs at all. This shows that, contrary to popular belief, increasing cleaning resources does not automatically result in a clean city.

"We see that the subject of the costs of cleaning stands out every time. Many managers do not have access to data such as agent's hourly costs", states Hervé Guillaume, Secretary General of the AVPU. However, the "bottom-up" approach requires this basic data, as well as that of the machines' hourly costs. Detailed knowledge of vehicles' hourly costs is the area of expertise of Hymexia, a company that advises communities in Switzerland on fleet management. This company has established statistics on the hourly costs of small sweepers (excluding staff), the structure of which is as follows:

Table 1: sweeper's hourly costs in euros (data © Hymexia)

The investment costs represent half the hourly cost (excluding staff costs), which is high and calls for intensive use of these resources. Still according to the same data, the operating time varies from 220 to 1,500 hours per year, with an average of 800 hours. The lifespan for an optimum rate of use is either 6,000 to 7,000 hours or 7 to 8 years. Maintenance costs at the end of this period are on average three times higher than at the beginning. Then the costs dramatically rise as the machines are often unavailable due to breakdowns or maintenance. According to Aurélien Gogniat, Director of Hymexia: "All the measures that make it possible to improve the efficiency of the resources committed have a great impact on hourly costs".

The top-down approach: an assessment of the resources allocated

In our previous article, we proposed a method of harmonising public accounts making it possible to compare the costs of urban cleaning. We collaborated with two Swiss cities of around 150,000 inhabitants or more, as well as with Mr Florian Rouland, business economics student at the HEIG-VD (University of Engineering and Management of the Canton of Vaud), to analyse their respective accounts in detail, then to restate the accounts according to the proposed framework. This study made it possible on the one hand to validate the proposed method (and to confirm the difficulties of analysing the accounts to determine the costs of cleaning) and on the other hand, to provide reliable results using a controlled approach.

This analysis confirmed that the practices were different between these two cities. For example, one of them chose, as the criterion for allocating costs, the type of surface, hard or black versus soft or green. In this case, the costs of cleaning the parks and maintaining the trees in the streets are allocated to the green spaces department. Cities have other choices of criteria to define the scope between their various services, such as processes and professions, or even the administrative organisation chart. These are generally internal criteria of the city, which are not perceived by users (a passer-by will be concerned by the dirt around a tree, not by the department which must clean it). In addition, definitions vary from city to city and sometimes result from historical choices.

In the approach we have proposed, we have favoured the "visible" perception of the user. In this case, the costs of cleaning green areas are attributed to the road network department, even if the tasks are performed by the green spaces department.

The partitioning of accounts, their level of detail and groupings by major accounts, re-invoicing between departments and special credits represented additional difficulties in allocating specific expenditure to each activity. The margins of error for restating city accounts to the proposed model are of the order of 10%. The amounts have been converted at a rate of 1.1 Swiss francs for 1 euro.

Table 2: cleanliness costs of the two Swiss cities analysed, converted into euros (CHF)

These results can be compared reliably with those of Zurich, given that this city uses an equivalent perimeter (in particular when taking into account wintering and green spaces). The costs of cleaning are 2.5 to 3.1 times lower. The number of inhabitants per road worker for cities A and B is quite close to the French average, while the city of Zurich can serve twice as many inhabitants with its staff. It should also be noted that the Zurich road network department provides services for third parties, such as the canton, car parks and surfaces of private companies, which allows 28% of its costs to be re-invoiced in the form of services.

Large differences in cleaning costs between cities

In a survey of 35 French cities, the AVPU estimates the costs of urban cleaning at €56 per capita per year. An analysis based on the detailed accounts of three Swiss cities arrives at an average of €89 per capita per year. Further research should be carried out to determine to what extent the difference between these results stems from the method and scope taken into account or from higher real costs, for example linked to the salary level in Switzerland.

On the other hand, it is very interesting to note that the costs of two of the Swiss cities are 2.5 to 3 times higher than those of the city of Zurich, and that in this case the results are entirely comparable, because obtained from the city accounting for an identical perimeter.

Yet the city of Zurich has the highest wage levels in Switzerland. It is therefore possible to have a clean city and attractive working conditions for employees, through efficient management of resources.

This difference of a factor of four on the scale of per capita costs is found for all the countries considered, regardless of the method used. This difference represents several million or tens of millions of euros depending on the size of the city. The causes are often attributed to the pressure on urban space; the propensity to dirty can depend on the number of non-resident users (commuters, tourists), behaviour, culture or other factors on which the city does not have control. Certainly, some external factors are outside the scope of action of the road network departments. However, this point of view does not do justice to efficient road network departments, which are equipped with management tools in order to control cleanliness and costs.

In our article on urban cleaning management, we noted the complexity of the tasks for the road network department, which must manage investments, prevention and cleaning under constant budgetary pressure (its budget does not increase with the growth of the population or the more intensive use of urban centres). How is it possible to manage this complexity without objective and quantified indicators?

We believe that the efficient use of resources is a more important lever for controlling costs and cleanliness. As proof, in our comparison of Swiss cities, the city of Zurich introduced the measurement of the level of cleanliness more than 20 years ago to control the quality of its services. Zurich is consistently ranked among the cleanest cities in Europe, with the highest satisfaction ratings in quality of life surveys. To our knowledge, it is also the only road network department that naturally has the relevant accounting information in relation to cleanliness.

What can't be measured doesn't get better. The level of cleanliness has a real impact on the quality of life of city users, as well as on the preservation of its environment. The costs of cleaning represent an important item in the budget of cities. Integrating accounting information and performance indicators for cleanliness and environmental impact therefore appear to us to be an essential step towards an attractive and sustainable city. Once these indicators have been established, it becomes possible to draw inspiration from the best performing city optimisation methods and observe the savings made with each iteration. We will present this to you in an upcoming Clean City Lab article.