7.7 Model for assessing the impact of city parks on socio-economic development and ESG (Ecology, Social, Governance) indicators

7.7.1 Purpose of the model

The model for assessing the impact of city parks on indicators of socio-economic development of the city and on ESG (Ecological, Social, Governance) has several key tasks:

Determination of environmental impact - Assessment of the ability of parks to reduce air and environmental pollution, improve water balance and urban biodiversity. Assessment of social impact - Measuring the degree of influence of parks on the health and well-being of citizens, improving the quality of life, providing access to green spaces for recreation and sports, creating conditions for communication and socialization of people. Management efficiency analysis - How effectively park management is carried out, whether there are clear rules and norms, how much funds are invested in the maintenance and improvement of parks. Assess the impact on the socio-economic development of the city - Does the presence of parks contribute to attracting tourists and creating new jobs, does it increase the quality of life and real estate value near parks, does it increase the city’s attractiveness for investment.

Such a model allows city leaders to make informed decisions about investments in parks and balance the interests of social, environmental, and economic goals.

7.7.2 Basic view of the model

The basic view of the Model for assessing the influence of city parks on indicators of socio-economic development of the city and on ESG (Ecological, Social, Governance):

\[ \frac{d Q_{i}}{d t}=k_{t} F_{i S}\left(\frac{d S}{d t}\right)+k_{t} F_{i V}\left(\frac{d V}{d t}\right)+k_{t} F_{i C_{k}}\left(\frac{d C_{k}}{d t}\right)+k_{t} F_{i C_{k}}\left(\frac{d C_{k}}{d t}\right) \ldots \ \ (1) \]

where \(\frac{d Q_{i}}{d t}\) - increment of the output indicator, \(F_{a b}\) - transfer function as increment b passes into increment a, \(k_{t}\) - normalization correction coefficient, \(\frac{d S}{d t}\) - change in the area of parks in the city created by the project, \(\frac{d V}{d t}\) - change in the volume of funds for greening in the city created by the project, \(\frac{d C_{k}}{d t}\) - inertial predicted changes in other indicators on which the desired indicator depends.

The main task of the Model is to find the transfer functions \(F_{a b}\).

7.7.2.1 Procedure for forming transfer functions

  1. General view of the transfer function.

\[ \begin{gathered} F_{\frac{d S}{d t}}=k_{1}\frac{1}{1+\frac{d S}{d t}}+k_{2} \frac{d S}{d t}+k_{3}\left(\frac{d S}{d t}\right)^{2}+k_{4} \frac{1}{1+\left(\frac{d S}{d t}\right)^{2}}+k_{5}\left(\frac{d S}{d t}\right)^{3}+k_{6} \frac{1}{1+\left(\frac{d S}{d t}\right)^{3}} \\ +k_{7} \frac{1}{1+\sin \left(\frac{d S}{d t}\right)} \\ +k_{8} \sin \left(\frac{d S}{d t}\right)+k_{9} \frac{1}{1+\sin \left(\frac{d S}{d t}\right)^{2}}+k_{10} \sin \left(\frac{d S}{d t}\right)^{2}+k_{11} \frac{1}{1+\cos \left(\frac{d S}{d t}\right)} \\ +k_{12} \cos \left(\frac{d S}{d t}\right)+k_{13} \frac{1}{1+\cos \left(\frac{d S}{d t}\right)^{2}}+k_{14} \cos \left(\frac{d S}{d t}\right)^{2} \end{gathered} \ \ (2) \]

We do not know what the character of the transfer function is, so we use an exhaustive list of functions and, with the help of training on a large sample, adjust the coefficients. Thereby we find the actual character of the function’s behavior.

  1. Formation of the training sample

The model uses several sets of data from various sources (full list of indicators Appendix 1 to this section):

  • 2GIS - area characteristics of SEO (Socio-Economic Objects),
  • municipal statistics of Rosstat,
  • approved urban planning standards (MNGP/RNGP)

To obtain a sufficient amount of data for training, we group cities by the principle of similarity; we called them “sister cities”. The principle of similarity in this context means the proximity of cities in a 400-dimensional (by the number of indicators available for each city) space. From this, a hierarchical classification is formed:

Example of city classification

We proceed from the assumption that the dynamics and effects are similar within the group. Thus, the training sample increases by 4-5 times.

  1. Data preparation

Since we are considering increments, it is necessary to subtract the trend from the original series so that the model takes into account only medium-term and short-term changes. We define the trend as the lowest frequency harmonic in the Hilbert transform. After which we get a series that is normalized. Thus we get a training sample.

  1. Compilation of a sample of dependent and influencing indicators

For each output indicator, indicators are determined that mainly influence the output indicator. To correct this sample, the completeness of the fraction of the variance of the original indicator is used, which is described by a set of \(n\) influencing indicators.

This happens based on the multiplication of PCA (Principal Component Analysis) matrices (synthetic variables are taken that together give 95% completeness) by its inverse, and the mutual influence of each indicator on each indicator is obtained.

For example, for the “Housing provision” indicator, this list includes:

  • Average monthly nominal wages of employees of organizations, total
  • Presence in the urban district (municipal district) of an approved master plan of the urban district (territorial planning scheme of the municipal district)
  • Budget expenditures for the maintenance and development of green spaces
  • Population as of January 1 of the current year
  • Average headcount of employees of organizations, total
  • Population younger than working age
  • Population arrivals (migration)
  • Population departures (migration)
  • Commissioning of individual residential buildings
  • Commissioning of residential buildings (apartment buildings and individual housing)
  • Number of births
  • Provision with sports facilities
  • Population of working age
  • Area of parks, sq. m.
  1. After obtaining the training sample, the model is calibrated by the mean absolute error for all coefficients. After the convergence of the model (error less than 10%), the model is considered trained and ready for use.

7.7.3 Input data

To assess the impact of investments in city parks, input parameters are used, the values of which are set by the user:

Name Unit of measurement Indicator code or variable name Data type Note
1 City location string The user selects one city from the list, the calculation module receives the 8-digit OKTMO code of the city
2 Project type type string The user selects one of two values: 1. Creation of a new object = “new” 2. Reconstruction of an existing object = “req”
3 Object area hectare C1188 floating point number A positive number from 0.1 to 1,000,000 is indicated
4 Investment volume million rubles C1115 floating point number A positive number from 0.1 to 1,000,000 is indicated
5 Project start year year start_year number An integer in the range from 2020 to 2040 is indicated
6 Year of completion of the investment stage of the project (transition to the operational stage) year invest_end_year number An integer in the range from the next after the start year of the project to the end year of the project evaluation period is indicated
7 Year of completion of the project evaluation period year end_year number An integer in the range from 2020 to 2040 is indicated. The year of completion of the project evaluation period must be equal to or greater than the year of completion of the investment stage of the project

Restrictions on input parameters applied in the Module for the influence of investments in city parks: - Minimum project start year - 2020 - Maximum year of completion of the project evaluation period - 2040 - Minimum project duration - 1 year - Maximum project duration - 20 years

7.7.4 Output data

As output parameters, the Module for the influence of investments in city parks forms the following set of indicators:

Group Name Unit of measure Indicator code
1 E (environmental) Total amount of pollutants tons C108
2 S (social) Captured and neutralized air pollutants tons C109
3 Average monthly nominal wages of employees of organizations rubles/month C261
4 Housing provision sq.m/person C804
5 G (governance) Number of small and medium-sized businesses per 10 thousand people units C1173
6 Migration growth of the population people С037

Properties of output indicators 1. Indicators are provided as a data series from the project start year to the project evaluation period end year. 2. The data series is an instantaneous effect for each year. 3. Effects from the object area (C1188) occur after the completion of the investment phase of the project, i.e., after the opening of the city park. 4. Effects from the volume of investments (C1115) occur immediately after the start of work.

7.7.5 Result of model application

The model is implemented in two versions:

During trial operation, the model confirmed its efficiency in target cities entered into the unified calculation Module for assessing the impact of investment projects on various socio-economic indicators of the city.

7.7.6 Requirements for model support and regularity of recalibration

It is advisable to carry out model calibration once a year after the publication of new statistical data used by the model. Increasing the amount of statistical data can improve the accuracy of the effects provided.

7.7.7 Assumptions necessary for carrying out calculations

Restoring missing data using “sister cities” leads to similar effects in the model output.

7.7.8 Scope (boundaries) of permissible application of the mathematical model

Since the Model is calculated on data of short-medium term increments and the influence of the trend is excluded. Investment projects with a park area of more than 30% of the total area of parks in the city and/or an investment volume of more than 50% of the total costs for improvement and greening of territories may lead to non-interpretable results. This situation is partially regulated by taking into account the investment potentials of the city. If the volume of investments exceeds the investment potential, then they are limited by its size.

7.7.9 Appendix 1

Code Name
C707 Gross value added of small and medium-sized businesses (SME)
C241 Payroll of all employees of organizations, total
C475 Revenue (For the reporting period)
C698 Taxable money income of individuals and individual entrepreneurs
C708 Tax revenues of budgets of all levels
C709 Local budget revenues (tax and non-tax)
C035 Number of deaths
C477 Cost of sales (For the reporting period)
C421 Inventories (At the reporting date of the reporting period)
C908 Emitted air pollutants
C261 Average monthly nominal wages of employees of organizations, total
C479 Gross profit (loss) (For the reporting period)
C640 Payments by organizations, including those related to employee compensation (For the reporting year)
C149 Total floor area of residential premises
C081 City budget expenditures, including: Social policy
C077 City budget expenditures, including: Education
C1173 Number of small and medium-sized businesses per 10 thousand people
C1156 Share of the length of public roads of local importance that do not meet regulatory requirements in the total length of public roads of local importance
C1163 Share of children aged 1–6 years registered for placement in municipal preschool educational institutions in the total number of children aged 1–6 years
C433 Current assets of organizations under Section II of the balance sheet (At the reporting date of the reporting period)
C1165 Specific value of thermal energy consumption by municipal budget institutions per 1 sq. m of total area
C1171 Presence in the urban district (municipal district) of an approved master plan of the urban district (territorial planning scheme of the municipal district)
C1174 Expenditures of the municipal budget for the maintenance of local government employees per resident of the municipality (2008, thousand rubles)
C050 City budget revenues, including: Personal income tax
C013 Population as of January 1 of the current year
C221 Average headcount of employees of organizations, total
C804 Housing provision (sq.m./person)
C1188 Area of parks, sq. m.
C040 Population younger than working age
C042 Population older than working age
C037 Population arrivals (migration)
C038 Population departures (migration)
C818 Number of employees in small and medium-sized businesses
C181 Number of public (general access) libraries
C757 Integral provision of SEO (Socio-Economic Objects)
C711 Full cost
C097 Commissioning of individual residential buildings
C697 Money income of the population
C096 Commissioning of residential buildings (apartment buildings and individual housing)
C815 Population employment
C002 Life expectancy at birth
C034 Number of births
C207 Number of shops
C516 Total financial result of the period (For the same period of the previous year)
C510 Net profit (loss) (For the same period of the previous year)
C1029 Retail trade turnover per capita
C1034 Number of small and micro enterprises per capita
C108 Total amount of pollutants
C753 Provision with sports facilities
C041 Population of working age
C109 Captured and neutralized air pollutants
C165 Number of hospital beds (round-the-clock hospitals)
C1115 Budget expenditures for the maintenance and development of green spaces
C699 Population income: social and other payments

Back to Model start

Back to chapter start