1 Introduction

Most cities are currently struggling to ensure accessible and efficient transport in their areas. The challenges planners face result from the negative effects of traffic congestion, mainly due to the expansion of individual transport. Efficient public transport is the basic tool for changing residents’ commuting behaviour and reducing the traffic of private vehicles and the resulting negative external costs, thus optimising social benefits.

The above-mentioned transport problems of urban agglomerations overlapped with, and were aggravated by, the ongoing COVID-19 pandemic, which impacted many areas of the economy, both in Poland and in the rest of the world. This article contributes to a growing body of research investigating the economic impacts of this pandemic. To provide some answers, this study undertakes a comparative analysis of statistical data for most provincial cities in Poland. This study addresses the need to understand the long-term impact of the pandemic on public transport services, which remains unclear and requires further research.

The novelty of this article is examining the correlation of changes in public transport ridership in individual Polish cities caused by the COVID-19 pandemic with the presented range of factors and conducted longitudinal studies. Multiple regression analysis was made possible by collecting data describing various aspects of public transport and populations. The advantage of the collected data is that they cover the entire analysed population, which eliminates the disadvantages of other sample-based analyses presented in the literature. On the other hand, the collected data do not allow for the differentiation of the profile of the analysed population. However, an attempt at such differentiation was made based on their comparison with the analysis of Central Statistical Office of Poland (CSO) data.

The structure of the text is as follows: Sect. 2 presents the transport problems of urban agglomerations and a review of the literature. Subsequently, in Sect. 3, the data and methods are explained. The key results from the analysis are presented in Sect. 4. In the final section, a summary of the results and their implications for practitioners is provided.

2 Transport problems of urban agglomerations

Technical progress and the increase in the wealth of society means that the level of motorisation is growing much faster than the population’s income. As a result, the car has ceased to be a luxury good and has become a common good; as an example, Poland has 656 passenger cars per 1000 residents (GUS 2021). In practice, it can be assumed that each household has at least one car. As a result, the price elasticity of the demand for public transport services has also changed. Previously, the demand for public transport was inelastic, as many consumers had no alternative. However, the popularisation of individual transport has made the use of public transport a consumer’s choice (Hebel et al. 2017, p. 119). In EU countries, short-distance public transport is characterised by an average price elasticity of demand of − 0.4 (Holmgren 2013, p. 105; Paulley et al. 2006, p. 295); similar results were also obtained for public transport in Poland (Hebel et al. 2017; Bernat 2008; Dyr 2006).

With the increase in the number of cars per household, road traffic congestion has become a major urban transport problem (Cervero 1998; Downs 1992). The dynamic development of individual motorisation entails significant external costs borne by the public, including road accidents, environmental pollution and the occupancy of valuable urban space (Nosal et al. 2010; Adler et al. 2016; Aftabuzzaman et al. 2010; Litman 2023). Parking fees in effect charge private car users with these external costs, while limiting the traffic of private vehicles and privileging public transport is a factor limiting these costs (Koźlak 2008, p. 122). Moreover, collective urban transport has a significant positive impact on the urban structure of agglomerations and their surroundings (Gzell 2010).

Increasing travel demand will compound the problem if appropriate solutions are not actively sought. Efficient public transport can be one of the potential solutions to the problem of urban road traffic congestion (Hyman et al. 2002; Pucher et al. 2007; Vuchic 1999). Public transport is beginning to be widely perceived by residents as an attractive alternative to the passenger car and contributes to refraining from using a car despite owning one (Holmgren 2013). This trend of using alternative transportation has been confirmed by studies conducted in the largest cities in Germany, Austria and Switzerland, which have shown significant reductions in car trips over the past 25 years despite high motorisation rates (Buehler et al. 2017). In the last 10 years, the number of registered cars per capita in Poland has increased by 47%; yet, the number of journeys made by public transport has also increased, albeit slightly, by 2% (GUS 2021). The necessary condition for the desired substitution of private car usage with public transport is its quality. Many factors influence the quality of public transport, as perceived by users (Common 1998). The evaluation of most qualitative factors is relative and difficult to quantify. However, the available results of several surveys conducted in recent years in various agglomerations in Poland show that, for passengers, the most important considerations are predictability and total door-to-door travel time by public transport (Mikulska et al. 2015; Kowalik et al. 2016; Milenkiewicz et al. 2011; Zajfert 2018).

The above-mentioned transport problems of urban agglomerations were heightened by the expansion of the COVID-19 coronavirus pandemic. The pandemic affected the global economy and societies on a scale unprecedented since the Great Depression (1929–1933) and is epidemically comparable to the Spanish flu of 1918 (Barro et al. 2020; Laing 2020). Public transport, used by millions of people every day, plays a significant role in the spread of communicable diseases among users (Buja et al. 2020; Muller et al. 2020; Musselwhite et al. 2020; Troko et al. 2011). As human contact exacerbated the spread of the virus, unprecedented restrictions on the organisation of public events and travel were introduced in many countries around the world (De Vos 2020; Tian et al. 2020).

Researchers have used one of two approaches to studying the issue of public transport from the perspective of the COVID-19 pandemic. On the one hand, scientists are studying the impacts of the transport sector on the spread of COVID-19. On the other hand, they are studying the impact of the COVID-19 pandemic, accompanied by government restrictions and public concerns, on decreases in travel demand levels of public transport.

The positive image of public transport, created with great organisational and financial effort, and the gradually achieved changes in the transport behaviours of city residents were significantly undermined by the outbreak of the pandemic, which completely changed the way urban agglomerations function as well as the transport problems they face. The results of several studies indicate that the ongoing pandemic significantly reduced people’s mobility and changed their commuting behaviours (Atchison et al. 2020; Bounie et al. 2020; Engle et al. 2020; Klein et al. 2020; Meier et al. 2020). Numerous studies conducted in many countries indicate that the change in commuting behaviours—largely favouring private cars—was mainly due to the fear of infection when using public transport (Eisenmann et al. 2021; Gutiérrez et al. 2020; Molloy et al. 2020; Shibayama et al. 2021). In one of the Spanish cities studied, during the initial period of the pandemic, the overall mobility of the population decreased by 76% and the decrease in public transport ridership was as much as 93% (Aloi et al. 2020). During the pandemic, the reluctance of passengers to ride in crowded vehicles was much higher than reported in the literature before 2020 (Tirachini et al. 2020).

This article contributes to a growing body of research investigating the economic impacts of the COVID-19 pandemic and focuses on the impact of the pandemic on the demand for public transport services. The COVID-19 pandemic in Poland has so far been analysed in the short, initial period of its development—5 months in 2020. It showed various strengths of its effects across voivodships. The results indicate that the forced lockdown to contain the development of the COVID-19 pandemic has effectively contributed to social distancing in public transport in Poland and that government restrictions, rather than a local epidemic status, induced a greater decrease in mobility (Wielechowski et al. 2020).

The results of other studies show that the travel demand pattern during COVID-19 indicates not only lower demand but less pronounced peak demand periods corresponding to commuting to and from work (Liu et al. 2020). The pandemic affected the daily mobility of various sociodemographic groups in different ways due to their different levels of exposure and perceived chance of contracting COVID-19. The drop in public transport usage was smallest for those who did not change their work-related behaviours, especially blue-collar workers. According to research, the total travel time of the latter decreased only by approximately 50%, which was mainly due to reductions in travel for recreational and shopping purposes (Matson et al. 2023). On the other hand, the total travel time of school students had reduced by up to 80%, mainly due to the introduction of distance learning (Borkowski et al. 2021).

The literature suggests that the number of passengers declined because some people were working from home in various countries, which reduced commuting. On the other hand, mode switches to cycling and driving at the expense of transit use were also observed in various countries (Follmer et al. 2020).

Travel demand was undoubtedly significantly reduced in the short term, but the decline rate varied across cities, which calls for an explanation. One factor that may explain the apparent variations in the decline in public transport ridership as a result of the ongoing COVID-19 epidemic is the different motivations of residents. During the pandemic, many residents who had previously chosen public transport despite owning a car changed their travel preferences out of concern for their health (Voß et al. 2020). It seems that this phenomenon should apply particularly to cities characterised by high-quality public transport services, as the substitution of collective for individual transport was the strongest in these cities (Voß et al. 2020).

Radical drops in public transport use resulted in significant reductions in organisers’ incomes from the sale of tickets. Although the proceeds from tickets only partially cover the operation of public transport, the significant reduction as a result of the pandemic while maintaining the current level of subsidies made it much more difficult to balance operating costs. At the same time, decreases in the number of passengers undermined the justification for maintaining the normal frequencies of many connections. During the pandemic, in many cities around the world, it became common practice to limit the number of public transport journeys or the number of stations served (DeWeese et al. 2020; TfL 2020; WMATA 2020; UITP 2020).

Studies conducted in Taipei indicate a highly heterogeneous decrease in metro usage under the influence of COVID-19, both spatially and temporally (Mützel et al. 2022). Notably, ridership decreased significantly, and the lower level seemed to persist even after COVID-19 was completely controlled. This trend poses a serious threat to public transit operators due to financial losses and represents a risk for the transition to more sustainable and public transit-based cities (Mützel et al. 2022).

3 Data and methods

3.1 The COVID-19 pandemic in 2020

Unit data on the size of public transport systems (public transport routes length, vehicle kilometres travelled, ridership, revenues from ticket sales and other parameters) were collected from the organisers of public transport in the voivodship capital cities in Poland. All 2020 data were compared with the 2018–2019 average (i.e. before the COVID-19 pandemic). Although these cities represent the largest urban agglomerations located in different parts of the country, the sample omitted other cities due to unavailable data. The selection of indicators describing public transport was also significantly limited by the availability of data.

The ongoing pandemic has contributed to a significant decline in public transport usage in all cities, but the magnitude of this decline varies. Table 1 summarises the data regarding this phenomenon. Individual voivodship cities are ranked according to population size. Overall, 17 cities are considered throughout this paper (incl. Table 1).

Table 1 Changes in public transport in voivodship capital cities in 2020
Full size table

As previously indicated, for passengers, the most important consideration for transport quality is total door-to-door travel time, including the time used to reach a stop and wait for the vehicle. The qualitative factors describing travel time have different values for different cities; also, these values changed to different extents during the pandemic as a result of the reduction of the service in that period. The range of available parameters describing the quality of public transport is limited and covers the technical characteristics on the supply side.

The public transport network in each city is differently structured. The network structure affects the time needed to reach the nearest stop and thus universal access for all residents. Therefore, one of the factors considered is the density of the public transport network (i.e. the length of the routes served per km2 of the city area).

One of the key qualitative factors that passengers consider is the time spent waiting for a public transport vehicle. At the same time, the structure of the public transport network and the length of common sections of individual communication lines affect the number of changes necessary during the journey, generating the need to wait for another vehicle. Therefore, the best available measure of the quality of this factor is the frequency of departures of all lines running on an average section of the network. This frequency increases in line with increases in the frequency of each line as well as the share of common sections of routes of individual lines. During calculations, it was assumed that, on a given route, all lines are served with the same frequency on all days of the week from 5 a.m. to 11 p.m. This assumption means it is possible to estimate the average time interval between subsequent departures of public transport vehicles on a given route.

The collected data indicate that certain cities in Poland, similar to other cities around the world, decided to reduce the costs of public transport by reducing the number of public transport journeys. However, this decrease worsened the perceived quality of public transport and may have resulted in further reductions in ridership in those cities. Therefore, another considered factor was the change in the supply of public transport services. They are presented in Table 1 (length of public transport routes and operational performance) and Table 2 (average frequency).

Table 2 Factors influencing the total travel time in voivodship cities in 2020
Full size table

One of the effects of the pandemic is the change in the form of work in many professions. Thanks to the use of popular information and communication technology, many types of work have taken their remote forms to a greater extent, including partial distance learning at universities. The adoption of telecommuting was one of the factors contributing to the reduction of private vehicle traffic on the streets in conjunction with other restrictions (e.g. public events). Consequently, less traffic has resulted in an increase in the driving speeds in cities. Long-term analyses of the data of navigation applications used while driving a car make it possible to determine changes in the average speed in individual cities in Poland. In Poland in 2020, the average travel speed increased by over 5% from 27.9 to 29.4 km more an hour (see Table 2). Thus, decreased congestion and travel time could encourage people who have so far chosen public transport due to traffic congestion to use their own car.

Table 2 summarises the data describing the change in the quality of public transport during the pandemic in voivodship cities.

Other reasons for the aforementioned disparities may be the different population sizes of individual cities or the employment structure of their residents. Expectedly, in smaller cities, a relatively large number of jobs are related to direct production or trade, which were not considerably reduced during the epidemic. At the same time, in large agglomerations, more workplaces are offices with jobs that are more susceptible to modification into a remote form, which has become common practice during the epidemic.

Detailed data for each voivodship city have been determined based on the database of the CSO of Poland: population size, number of pupils, students and employees, average salary, number of employees with tertiary education, and number of employees in industry sectors (i.e. industry, construction, trade, repair of motor vehicles, transport and warehouse management). Table 3 presents the collected data describing the residents of voivodship capital cities in 2020.

Table 3 Characteristics of voivodship cities in 2020
Full size table

Data on the number of employees who could work remotely during the pandemic were significantly limited due to the aggregation of the available database of the CSO. Therefore, the best approximation was the share of students and employees estimated in two different ways. For the first method, it was assumed that remote work most often concerns employees with higher education, so their share was calculated. In the second method, the share of employees in sectors enabling remote work and sectors whose functioning may have been limited as a result of the pandemic was calculated (i.e. all sectors except industry, construction, trade, car repair, transport and warehouse management). In a further analysis, it was assumed that all students, trainees and employees of those groups ceased regular travel by public transport during the pandemic. Of course, this assumption is not true, but it is an acceptable approximation when comparing changes in demand in individual cities.

Another factor explaining the aforementioned disparities observed in demand may be the wealth of the local community and the residents’ purchasing power for public transport services. This variable was defined as the number of journeys that can be purchased for an average gross salary in a given city in 2020. The level of the average salary for individual cities was determined based on the CSO database. The cost of one trip by public transport was calculated according to the actual costs of an average trip. Due to the complex and varied fare systems in individual cities, this should not be equated with the price of a single ticket for public transport. Therefore, the trip cost was calculated as the quotient of the total revenues from ticket sales and the number of journeys in the city in 2020.

In total, we collected data describing 11 factors that may explain the change in demand for public transport services during the COVID-19 pandemic:

  1. 1.

    Changes in the length of public transport routes between 2020 and the 2018–2019 average

  2. 2.

    Changes in the operational performance (vehicle—km) between 2020 and the 2018–2019 average

  3. 3.

    Changes in the frequency between 2020 and the 2018–2019 average

  4. 4.

    Changes in the average speed of private cars between 2020 and the 2018–2019 average

  5. 5.

    Purchasing power of public transport services in 2020

  6. 6.

    Population in 2020

  7. 7.

    Density of the public transport network (km/km.2)

  8. 8.

    Changes in the length of the public transport network (km) between 2020 and the 2018–2019 average

  9. 9.

    Average frequency in 2020 (minutes between successive departures)

  10. 10.

    Estimated share of groups avoiding public transport during the COVID-19 pandemic among passengers who use it regularly—students and employees of selected sectors

  11. 11.

    Estimated share of groups avoiding public transport during the COVID-19 pandemic among passengers who use it regularly—students and employees with higher education

The relationship between the change in demand for public transport services and all the collected determining factors was modelled using multiple linear regressions. This analysis was performed for all listed explanatory variables and did not indicate extreme values of any of the observations. However, for the last six explanatory variables (6–11 in the list above), the P-value exceeded 5%, which is why they were considered statistically insignificant and eliminated in the next iteration of the analysis. Tables 4 and 5 present the results of this multiple regression analysis.

Table 4 The multiple regression analysis of changes in public transport ridership with five explanatory factors
Full size table
Table 5 The multiple regression statistics
Full size table

Among the factors under the multiple regression analysis, the lowest standard error was found for the purchasing power of public transport services. The positive sign of the coefficient indicates that the largest decrease in public transport ridership took place in cities where residents had the lowest purchasing power.

The factor with the second-lowest standard error was the relative difference in the average speed of private cars between 2020 and the 2018–2019 average. The negative sign of the coefficient indicates that the largest decrease in public transport ridership took place in cities where the speed of passenger cars increased the most.

The next three explanatory factors describe the network and supply of public transport services. The decrease in public transport ridership was the smallest in cities where the length of routes increased the most, the average waiting time for a vehicle increased the least and transport performance was reduced to the smallest extent.

Table 6 shows the revenues from the sale of public transport tickets and the degree of coverage of operating costs in voivodship cities. The decline in the number of passengers as a consequence of the ongoing pandemic is reflected in those revenues. However, the changes in these two values in individual cities were not proportional. During the relevant period, the ticket price did not change in any of the analysed cities, indicating that the price cannot be a factor explaining these differences.

Table 6 Changes in the economic parameters of public transport in voivodship cities in 2020
Full size table

The collected data made it possible to compare the changes in revenues from ticket sales and the level of cost coverage in relation to all other factors mentioned above. These relationships were modelled using multiple linear regressions, which are presented in Tables 7 and 8).

Table 7 The multiple regression analysis of changes in revenues from ticket sales with three explanatory factors
Full size table
Table 8 The multiple regression statistics
Full size table

For all explanatory variables, the obtained P-value was higher than 5%; therefore, they were considered statistically insignificant. A significant P-value was obtained for only three factors, but it was about 15% for them. A subsequent multiple regression analysis of the three factors revealed that such a model explains only 32% of the observations (Adjusted R2), which confirms its uselessness.

3.2 Longitudinal study

For two public transport systems, the Public Transport Authority in Warsaw (ZTM Warszawa) and the Gdańsk Urban Railroad (SKM Trójmiasto), data on monthly ridership from January 2018 to May 2022 were available. This made it possible to conduct longitudinal studies. For each month in the years 2020–2022, the change in the number of travellers was determined and compared to the average for the same month in 2018–2019. The calculated monthly changes made it possible to create a chart overlaid with the chart of the number of new weekly COVID-19 cases in Poland, which is available on the Internet (Fig. 1). It was not possible to obtain data for the same population range. Therefore, data on COVID-19 cases cover the whole of Poland, while data on public transport in individual months were available for only two agglomerations. Despite this inconsistency, the analysis based on this chart seems to be methodologically correct in the case of Poland. This accuracy is likely due to the relatively small area of the country, its even population density and the fact that at no time during the pandemic were administrative bans introduced on moving around its area. The graph deliberately presents a decrease in transport (i.e. the greater the decrease, the higher the curve) because it illustrates the correlation (or lack thereof) with the increase in the number of COVID-19 cases. In the case of an increase in transport, the curve is below 0.

Fig. 1
figure 1

Comparison of the number of COVID-19 cases in Poland and changes in public transport ridership in Warszawa and Gdańsk. Left scale: daily number of new COVID-19 cases detected in Poland (7-day average), right scale: the decrease in public transport ridership compared to the same month in 2018–2019 Sources: COVID-19—https://www.worldometers.info/coronavirus/country/poland/#graph-cases-daily, public transport in Warsaw—https://www.ztm.waw.pl/statystyki/, railway transport: https://dane.utk.gov.pl/

Full size image

Comparing the same range of monthly data in a longitudinal study for two different agglomerations could indicate the long-term impact of a pandemic on the demand for public transport services. The 80% decrease in public transport ridership visible in the first months was not due to the current state of the epidemic. After more than a year of this pandemic, from June 2021, an increase in public transport ridership was visible compared to previous months. However, ridership was still approximately 20% lower than in the same month before the pandemic. From that moment, a significant difference in the commuting behaviours between Warsaw and Gdańsk is visible.

During the fourth wave of the pandemic (Nov 2021), no additional administrative restrictions on public spaces were introduced, and the number of passengers in public transport continued to increase. However, during the fifth wave of the pandemic (Jan 2022), a noticeable decrease of approximately 10% was reported.

4 Discussion

4.1 The COVID-19 pandemic in 2020

It is difficult to assess the demand side of non-economic aspects, which, as shown in Fig. 1, appear to play a key role in changing commuting behaviours in the face of the ongoing COVID-19 pandemic. Due to the lack of satisfactory results, an attempt was made to find other factors that could explain the disparities in the decline in the ridership levels of public transport, which were presented in Table 4.

The model created based on the multiple regression analysis of five factors explained 75% of the observations (Adjusted R2), the estimation error of this model was 13%. Among the factors under analysis, the purchasing power of public transport services was notable due to its high coefficient of determination. The obtained result indicates that the decline in public transport resulting from the ongoing COVID-19 pandemic was less severe in cities with lower real costs of travel for the average resident. This is consistent with the results of other studies indicating that the discrepancy in mobility depends on the level of income (Coven et al. 2020).

The second factor with a low standard error was the relative difference in the average speed of private cars between 2020 and the 2018–2019 average. This difference results from the increase in the attractiveness of individual transport in relation to public transport. The next three explanatory factors describe the network and supply of public transport services. In line with intuitive expectations, smaller reductions in supply during the COVID-19 pandemic resulted in smaller decreases in the ridership levels of public transport.

However, the economic condition of a city does not depend directly on the number of passengers using public transport. Revenues from the sale of public transport tickets have a much greater impact on the economic condition of individual cities, as it determines the degree of coverage of operating costs. Public transport providers must trade off between minimising the spread of COVID-19 and providing an affordable travel choice. Developed by Srivatsa Srinivas (2023), a strategic queueing model explains how pricing can influence the behaviour of commuters during COVID-19. According to it, the public transport provider should decrease the prices to filter out the commuters who can afford other modes of transport and provide services to the commuters who cannot afford other modes of transport. The decline in the number of passengers as a consequence of the ongoing pandemic has been reflected in revenues. However, the changes in these two values ​​in individual cities were not proportional (see Table 1). In Poland during the relevant period, the ticket price did not change in any of the analysed cities, so this price cannot be a factor explaining these differences. Fare systems favour long-term (e.g. monthly, quarterly) tickets; therefore, the variation in decreases in transport and revenues may indicate a change in the sale of a specific type of ticket, and thus a change in the nature of journeys. In the cities of Białystok, Rzeszów, Wrocław and Kraków, the changes in revenues from ticket sales were much lower than the changes in the numbers of passengers, which may indicate a significant decrease in the share of journeys with season tickets. Given the uncertainty of the pandemic’s duration, the abandonment of long-term period tickets seems to be an understandable phenomenon, which was also reflected in other studies (Jenelius et al. 2020).

In the cities of Zielona Góra and Szczecin, the decrease in the number of journeys was smaller than the decrease in revenues from ticket sales. In the statistical data, in the case of transfers made by a passenger, each stage of their journey was counted as a separate journey. Therefore, this difference in the size of the decrease may result from an increase in the number of transfers made during the journey, which may indicate changes in the layout of the public transport network or a greater limitation of trips made without transfers. The avoidance of transfers applies to a greater extent to elderly people, who may have reduced their use of public transport during the pandemic more than younger people.

The multiple regression analysis changes in ticket sales revenues and the level of cost coverage did not allow for reliable results; for all explanatory variables, the obtained P-value was higher than 5%. The available data did not allow us to pinpoint any factor that could explain these disproportions, which may be the result of a relatively small sample. The previously analysed changes in the demand for public transport services concerned only 17 cities, but they were the result of the commuting behaviours of many thousands of their residents. On the other hand, the change in economic results was due to the managerial skills of a relatively small group of employees managing public transport in individual cities as well as the individual characteristics of these cities.

4.2 Longitudinal study

The comparison of the weekly numbers of new COVID-19 cases and changes in transport in consecutive months, presented in Fig. 1, shows that these values are not correlated. The visible initial 80% decrease in public transport ridership was not due to the current state of the epidemic but rather the strict administrative restrictions that were introduced (Bonaccorsi et al. 2020; Liu et al. 2020; Pepe et al. 2020; Pullano et al. 2020; Queiroz et al. 2020; Wielechowski et al. 2020) and the fear of infection (Chan et al. 2020; Wang 2014). Over time, compliance with the restrictions turned out to be unstable (Kim et al. 2021); it changed depending on the available information about the COVID-19 virus itself and the level of risk aversion (Yuksel et al. 2020). The decrease in ridership stemmed primarily from a reduction in the number of active public transport travellers, while the daily average number of trips per active traveller remained relatively stable (Jenelius et al. 2020).

After more than a year of this pandemic, mass vaccination and the end of the third wave of infections (i.e. from June 2021), the psychological impact of COVID-19 on the commuting behaviours of residents decreased. During this period, most of the restrictions had already been lifted or significantly limited (e.g. school activities were fully restored). This translated into an increase in public transport ridership. As other studies showed, even after the easing of restrictions related to COVID-19, concerns about the level of congestion as well as the cleanliness and effectiveness of vehicle disinfection were much more important in urban residents’ choice of means of transport than before the outbreak of the pandemic (Eisenmann et al. 2021; Beck et al. 2021).

From that moment, a significant difference in the commuting behaviours between Warsaw and Gdańsk becomes visible. In the capital of Warsaw, ridership was still lower by approximately 25% compared to the corresponding period before the pandemic. Regarding regular journeys (i.e. based on long-term tickets), public transport ridership was lower by over 30%. The disproportion in sales of long-term tickets in relation to the total number of tickets visible in this period seems to result from residents’ uncertainty regarding the further development of the pandemic. As a result, season tickets were abandoned in favour of one-way tickets, which is in line with the results of other studies (Jenelius et al. 2020). On the other hand, this may also have been influenced by the then-approaching holiday season.

Gdańsk is located on the Baltic Sea and is a frequent destination for summer holidays. Therefore, the volume of ridership for this area already returned to the pre-pandemic level in the summer of 2021. The disproportion between the Gdańsk and Warsaw agglomerations is not explained by the estimated percentage of employees able to work remotely. The estimated share of students and employees of selected sectors is higher in Gdańsk, while the share of students and employees with higher education is higher in Warsaw (see Table 3).

During the fourth wave of the pandemic (Nov 2021), no additional administrative restrictions on public spaces were introduced, and the number of passengers in public transport continued to increase. For the first time since the outbreak of the pandemic, Warsaw’s PT passenger numbers were only 20% lower than in the same month prior to the pandemic. A noticeable decrease of approximately 10% was seen only during the fifth wave of the pandemic (Jan 2022). The disappearance of social fears and the almost complete abandonment of restrictions could explain the slight decline in public transport attendance during the last two waves of the virus from Nov 2021 to Feb 2022. This, in turn, may be one of the reasons why the number of new COVID-19 cases at that time was the highest since the beginning of the pandemic.

One of the effects of the pandemic is the change in the form of work in many professions. Thanks to the use of popular information and communications technology tools, many types of work have taken their remote forms to a greater extent, including partial distance learning at universities (Matson et. al. 2023). During the pandemic, it was one of the factors limiting the movement of private vehicles on the streets, which resulted in an increase in driving speeds in cities (see Table 2). This could encourage people who have so far chosen public transport due to traffic congestion to use their own car. Remote work and learning introduced during the pandemic was just as efficient as its previous forms in some sectors and fields, but cheaper. Therefore, a possible decrease in demand for public transport due to the continuation of remote work should be considered, which needs to be verified in future studies.

5 Summary

The key factor determining the choice of public transport rather than individual transport is the competitiveness of the total door-to-door travel time by public transport, including times for reaching a stop, waiting for a vehicle and the journey itself. Meanwhile, the positive image of public transport, created with great organisational and financial effort and the gradually achieved changes in the transport behaviour of residents were significantly undermined by the outbreak of the COVID-19 pandemic. This outbreak resulted in a significant decrease in public transport ridership. The ongoing pandemic has contributed to a significant decrease in transport in all cities, but the magnitude of this decrease varied. The analysis of the collected data shows that these discrepancies are related to public transport quality measures. However, to the greatest extent, the decrease in ridership levels of public transport due to the ongoing COVID-19 pandemic is less severe in cities with lower real travel costs. In addition, the decrease was less severe in cities where the attractiveness of alternative individual transport did not increase. The comparison between changes in ticket sales revenues and the level of cost coverage in individual cities as a result of the ongoing pandemic did not reveal any factor that could explain the apparent disproportions.

The magnitude of this decline in the following months in 2020 indicates that it is not correlated with the number of new cases of COVID-19 but rather a result of the administrative constraints imposed and the psychological impact of the development of the pandemic itself on the commuting behaviours of residents. In the second half of 2021, after the third wave, mass vaccination and the lifting or significant reduction of most of the restrictions, this impact was much lower. The disappearance of social fears and the almost complete abandonment of restrictions could explain the small decline in public transport attendance during the last wave of the virus between Nov 2021 and Feb 2022. This, in turn, may be one of the reasons why the number of new COVID-19 cases at that time was the highest since the beginning of the pandemic.

Compared to their traditional forms, remote work and distance learning introduced at that time in some sectors and fields were just as efficient but cheaper. Therefore, a possible decrease in demand for public transport due to the maintenance of remote work should be considered, which needs to be verified in future studies. Ridership may also be influenced by the concern for the cleanliness of public transport while assessing the quality of its services, which was not seen as a significant factor until the spread of the COVID-19 pandemic. The temptation of public transport operators to increase prices or reduce the supply of services to counteract declining financial performance could further discourage the use of public transport services.

This study focused on one country, but it can be assumed that the situation described may be typical for other countries with similar social structures, cultures and mobility behaviours. This assumption is confirmed by the fact that the obtained results are consistent with the conclusions of other studies carried out in other countries using different methods. Such support confirms the correctness of the obtained results and their universal character.

The presented results may be of great importance when developing a policy for strengthening public transport. These aspects are critical to achieving a sustainable transport system in the medium to long term, even looking beyond the coronavirus pandemic.