Working Conditions and Satisfaction of Transport Service Employees

Globalisation has opened up the finance, commodity and service markets. The opening of the economy is associated with lower costs of transport, communication and technology development; expansion of transnational corporations; and the growing interdependence of economies related to the new organisation of the value-added chain, among other things (Costinot, Vogel, & Wang, 2013, pp. 109–144). The main phenomena determining the trends in the observed changes are technological progress and the reduction of barriers to international trade.

Technological progress entails changes in the organisation of work. Originally, production was based on performing simple work (Bhagwati & Dehejia, 1993), but now, it has been replaced by methods characterised by high use of human capital. As a result of these changes, the importance of the knowledge, qualifications and competences of employees has increased. Due to technological progress, in the modern economy, workers have become carriers of capital, not labour, which was the case in traditional labour market models (Drucker, 1999). Technological progress also influences the labour market through routinisation (schematisation), modularisation (mass production of components) and standardisation (standardisation of processes) and, on the other hand, through task specialisation and the requirements of employers towards employees (Nicholson, 2004, p. 78). Each of these processes entails fundamental changes in the characteristics of contracts in the labour market, with the traditional form of contract being gradually replaced by flexible forms of employment, i.e. a specific task contract, contract of mandate, civil law contract, self-employment, and so on.

In the simplest terms, the employment contract has both economic and psychological aspects. The economic aspect is defined as all financial benefits resulting from the employment relationship, which are included in a contract. The psychological aspect can be understood as an unwritten set of mutual expectations between an employee, an organisation and its members. Thus, a contract in the labour market is a multidimensional concept. It entails a number of expectations and obligations that determine the satisfaction of an employee and an employer in a mutual relationship. Most of the psychological dimensions of an employment contract are immeasurable and therefore difficult to statistically document. There are, however, some basic frameworks within which the economic aspect of an employment contract can be described. The standard employment contract includes the following basic information (Liptak, 2011, pp. 1–13):

subject of labour (time or product);

In addition to the aforementioned formal dimensions of a contract, there are a number of others that determine the mutual satisfaction of the parties and thus explain the fact of the creation and durability of employment.

Dynamic globalisation processes make enterprises adapt in many aspects. The most important of them is the transformation of labour demand, where the size, structure and characteristics of this demand vary (Costinot, Vogel, & Wang, 2013, pp. 109–144). As a consequence of globalisation, an increased risk of doing business, mostly related to the increased volatility of the demand for the final goods, is noticeable. In the long run, this leads to a new understanding of the comparative advantages resulting from the possibility of adapting the labour resource as a result of heterogenisation of an employment contract. The new organisation of the labour market related to these processes contributes to its segmentation (Costinot, Vogel, & Wang, 2013, pp. 109–144). As a result of these changes, highly qualified specialists in narrow fields are sought. The work efficiency of specialists is particularly high if they perform activities strictly adequate to their competences. Another group of employees sought after in the labour market are those who can perform simple jobs and constitute a relatively homogeneous and low-paid workforce, usually without the possibility of promotion.

Flexible forms of employment have also developed dynamically in the labour market. They are associated with the transfer of risk to employees and the reduction of fixed production costs. This type of employment includes civil law contracts, remote and part-time work and hiring employees through temporary employment agencies (Liptak, 2011). Work organisation has also changed and involves increasing the share of part-time employment, flexible working hours, multitasking and the simultaneous implementation of several projects according to the market demand (Liptak, 2011, pp. 1–13). Employers’ needs for greater flexibility of labour resources have met with the technical development of the possibility of providing work in a non-standard way (e.g. remote work). This, in turn, has contributed to the activation of social groups previously excluded from the labour market (such as individuals with disabilities, the elderly and women on maternity leave), for whom undertaking work under traditional conditions was associated with an excessive opportunity cost. Therefore, making the forms of employment more flexible and diversified is associated with mutual benefits. At the same time, it serves to meet the needs of employers in terms of the flexibility of managing labour demand in order to maintain a competitive position in the modern economy and allows individuals who were previously professionally inactive to become more active.

The changes taking place in the economy also have negative consequences. They mainly concern the supply side of the labour market (Liptak, 2011). For a part of the workforce, the above-described processes are mostly associated with a decline in the security and stability of employment. This has important economic, as well as psychological and social, consequences. It should also be emphasised that making the employee–employer relationship more flexible also brings problems for employers. The most serious of them is the reduced effectiveness of investments in specific human capital, higher costs for recruiting employees and increased costs related to employment rotation.

The multiple forms of employment contracts are reflected in the development of economic thought. Almost all the progress in modelling the labour market in recent years concerns the heterogeneity of the parties to an employment contract. The search theory explains how heterogeneous employers search for heterogeneous employees, bearing the transaction costs of this process, and how job candidates look for suitable offers (Mortensen, 1986, pp. 849–919). In the classical approach, this heterogeneity only referred to the amount of productivity and remuneration. In later analyses, the search is presented as a whole set of relations characterising the costs and benefits for both parties (Sullivan, 2014, pp. 472–507).

In theoretical terms, the goal of the labour market is general social welfare (Jemielniak & Latusek, 2005). Increased heterogeneity of employment conditions may be beneficial to society, and an increased variety of forms of employment should lead to a situation in which it is possible to achieve equilibrium. The state of equilibrium will be reached only when the heterogeneous connections of employees and employers are characterised by such a composition of contractual terms that, on the one hand, it entails low costs from the employer's point of view, and on the other hand, it brings the highest possible level of job satisfaction derived by an employee. An increased overall level of welfare results from the fact that an increased variety of forms of employment theoretically increases the number of dimensions in which it is possible to optimise contracts.

In fact, there are problems with coordination due to the asymmetry of information between the demand and supply sides (Mortensen, 1986). As a consequence, information asymmetry may prevent optimal allocation from being achieved. The effects of the heterogenisation of employment conditions will also vary depending on the balance between supply and demand in the labour market. Depending on the general situation in the labour market, the changes may bring net benefits to employees or transfer the employer's risk to them.

Earlier, working time and remuneration were the basic attributes of employment. Currently, employees look for more attractive employment attributes in the labour market. One of the first studies in this area was conducted in 1976 by Hackman and Oldham using the job characteristic model (Hackman & Oldham, 1976, pp. 250–279). The study identified three mental states of an employee: a sense of meaningfulness (perceiving one's work as important and meaningful); a sense of responsibility; and knowledge of outcomes (or knowledge of work results). The listed states are evidence of job satisfaction and internal motivation. In order for the desired mental states to occur, employment should be characterised by key features, such as skill variety, task identity, task significance (the degree to which work influences the life or work of other people), autonomy (the degree of freedom with which a person has the ability to shape the way of performing tasks) and feedback (obtaining an assessment from institutions, superiors, colleagues or clients).

The empirical research confirms the influence of work characteristics on the sense of job satisfaction and the motivation of respondents. However, there are some methodological problems regarding the obvious correlation of job characteristics and a sense of job satisfaction.

The study entitled ‘The Demand for Leisure and Nonpecuniary Job Characteristics’ (Atrostic, 1982, pp. 428–440) describes the determinants of the declared utility derived from work by employees. Satisfaction with employment was measured by assessing the respondent's compliance with the statements regarding the benefits obtained from the work performed. The following employment attributes were considered: subjective assessment of satisfaction with the achieved professional position; training and professional development opportunities; satisfaction with adjusting work to employee expectations; degree of responsibility burden; degree of being subject to time pressure; complexity and difficulty of entrusted tasks; atmosphere in the team; and satisfaction with received remuneration. Moreover, 16 zero–one variables describing job characteristics were taken into account, which, according to the respondents, proved to be crucial in making the decision to take up their current employment. Among them, the following were distinguished: autonomy at work, prospects for financial success, amount of free time, employment stability, the possibility of helping others, compliance of the nature of work with the respondent's education, the possibility of working with people, interpersonal relations, remuneration including benefits, compliance of work with interests, the ability to face challenges, the possibility of continuing the family business and social status obtained through employment. The conducted study is distinguished by a wide range of attributes included in the analysis. It should be borne in mind that these measures were only based on responses relating to motivation at the time of making the employment decision.

The research conducted by Garcia and Molina (1999, pp. 189–204) takes into account the factor of occupational prestige as a determinant of utility derived from employment. Prestige is perceived in both social and personal aspects. In the social aspect, prestige results from the social position, and in the personal aspect, it is synonymous with recognition and respect. Prestige relates to a specific person and has its source in the way he/she fulfils a given social role and his/her personal characteristics. Currently, prestige is one of the basic factors motivating employees (Domański, 1999). The study used the method of multi-step estimation procedures. The authors analysed the processes of making professional decisions by Spanish workers. Decisions were made based on the offered remuneration and non-wage employment benefits. The study used a generalised linear expenditure model and the two-step Heckman correction. The following research stages were distinguished:

The selection of individuals was modelled to the working population (i.e. the group for whom information on remuneration was observed).

Consequently, the results of the research made it possible to calculate the financial value of compensation for worse working conditions.

The study was conducted using simple methods of mathematical statistics and regression models, where parameters are most often estimated using the least squares method. The variable related to job characteristics was mainly described by the determinants of earnings or declared job satisfaction. In these models, the dependent variable was most frequently wage per hour of working time (or its logarithm), and socioeconomic variables were among the independent variables. The most important problems of the research included the following: identification of the dependent variable, selection of independent variables and the lack of information about alternatives, and the respondent's situation at the time of making a decision about employment.

Scott used the conditional choice method for the first time in the study of preferences towards employment in 2001. He investigated preferences of nurses towards work depending on the characteristics of the offered employment (Scott, 2001, pp. 329–347). The conditional choice method involves formulating sets of hypothetical alternatives, where a respondent selects the most preferred answer. This method assumes that the ceteris paribus condition is met, along with the features describing the alternatives. The questionnaire survey solves the problem of variables omitted in the model because in this approach, the estimation of the parameters of the utility function derived from specific levels of attributes describes the alternatives presented to respondents. The aforementioned method was also used by Janicka (2013), who examined the decision to take up employment and analysed the determinants of employing immigrants in simple jobs and in particularly low-prestige professions. The author estimated the parameters using the random parameters logit (RPL) model and the multinomial logit (MNL) model. The RPL was used for the first time in Polish labour market research.

The above-mentioned method was also used by Gajderowicz in his empirical study entitled Korzyści z zatrudnienia. Dekompozycja i wycena [Benefits of Employment: Decomposition and Valuation] (Gajderowicz, 2016). The research process involved selecting attributes and constructing a form using qualitative research, conducting a pilot study and carrying out a study with the DCE method using a computer application on >800 students and graduates of social studies at Warsaw universities, aged 21–30 years (Gajderowicz, 2016, pp. 108–153). Using the DCE method, pecuniary estimates of the value of individual levels of work attributes were obtained. It also defined which features of employment are particularly important for potential employees and help them decide about taking up a job. The modelling covered, in particular, the preferences regarding non-wage benefits, features of the job and type of work, job prospects and the method of using working time. The utility function parameters were estimated using the MNL and RPL models. In the econometric analysis, the MNL model was used, where estimates of the parameters of the utility function derived from employment with particular characteristics were obtained. Moreover, the RPL model was used to obtain estimates of the parameters of their statistical distribution. Then, the differentiation of preferences towards employment in the sample was discussed. Finally, wage thresholds were calculated taking account of the non-wage features of employment; in other words, the individual levels of employment attributes were valued in financial terms. The research results show that decisions regarding work and the utility derived from it depend directly on the following: non-wage financial benefits provided by an employer, type of contract, method of payment of remuneration, benefits related to work performance and individual external benefits (development opportunities, prestige of the position held). Another category comprises the physical aspects of the performed work (working time, travel time, pleasure from the tasks performed), as well as the mental conditions of performing entrusted tasks (team atmosphere, as well as levels of stress and control). The study showed that the financial value of utility derived from individual employment characteristics was significant in relation to the average market wage during the study period. The valuation of some of the attribute levels exceeded half of the average market remuneration in the analysed period, which proves that non-wage streams of benefits can actually determine the utility derived from employment. To sum up, non-wage streams of benefits may actually decide about starting, maintaining and supplying work.

A review of research described in the literature reveals that the assessment of utility derived from work examined with the DCE method has practical applications. By determining the parameters of the utility function, it is possible to construct offers that bring a candidate a higher job utility at particular employment costs.

The -MNL model is the econometric tool used to estimate the parameters of the preference models (McFadden & Train, 2000). Modelling of preferences has its source in the random utility model, where consumers make choices between alternatives based on preferences included in the utility function in the following form (Czajkowski, 2008): Uji=β′xij+ɛij{U_{ji}} = \beta '{x_{ij}} + {\varepsilon _{ij}} where β is a vector of utility function parameters and x is a vector of consumer-relevant attributes i.

The utility can therefore be divided into the sum of the observable deterministic part V_nj = β * x_ij and the unobservable part ɛ_nj. The logit formula is based on the assumption that ɛ_nj has independent and identical distributions for each respondent and alternative. If ɛ meets the independent and identically distributed (IID) assumption, the difference between ɛ_nj and ɛ_nj is given by the following logistic distribution (Czajkowski, 2008): F(ɛnji*)=eɛnji*1+ɛnji*F\left( {\varepsilon _{nji}^*} \right)={{{e^{\varepsilon _{nji}^*}}} \over {1 + \varepsilon _{nji}^*}} The probability of choosing an alternative i by consumer n can be written as follows (Czajkowski, 2008): Pni=Prob(Vni+ɛni>Vnj+ɛnj ∀j≠i)=Prob(ɛnj<ɛni+Vni−Vnj ∀j≠i)\matrix{{{P_{ni}} = {Prob}\left( {{V_{ni}} + {\varepsilon _{ni}} > {V_{nj}} + {\varepsilon _{nj}}\,\forall j \ne i} \right)} \hfill \cr { = {Prob}\left( {{\varepsilon _{nj}} < {\varepsilon _{ni}} + {V_{ni}} - {V_{nj}}\,\forall j \ne i} \right)} \hfill \cr } It follows from the assumption about the independent random component that the cumulative distribution of this probability will be the sum of individual cumulative distributions: Pni|ɛni=∏e−e−(ɛni+Vni−Vnj){P_{ni}}|{\varepsilon _{ni}} = \prod {{e^{ - {e^{ - \left( {{\varepsilon _{ni}} + {V_{ni}} - {V_{nj}}} \right)}}}}} Since ɛ is unknown, the probability of choosing an alternative can be expressed as follows (Czajkowski, 2008): Pni=∫(∏e−e−(ɛni+Vni−Vnj))e−ɛnie−e−ɛnidɛni{P_{ni}} = \int {\left( {\prod {{e^{ - {e^{ - \left( {{\varepsilon _{ni}} + {V_{ni}} - {V_{nj}}} \right)}}}}} } \right){e^{ - {\varepsilon _{ni}}}}{e^{ - {e^{ - {\varepsilon _{ni}}}}}}d{\varepsilon _{ni}}} After multiple transformations, the above equation can be written as follows: Pni=eVni∑jeVnj{P_{ni}} = {{{e^{{V_{ni}}}}} \over {\sum\nolimits_j {{e^{{V_{nj}}}}} }} The probability that consumer i will choose alternative j from the J available alternatives can then be written as follows (Czajkowski, 2008): P(∀q≠jUij>Uiq)=exp(β′xij)∑q=1Jiexp(β′xiq)P\left( {{\forall _{q \ne j}}{U_{ij}} > {U_{iq}}} \right) = {{\exp \left( {\beta '{x_{ij}}} \right)} \over {\sum\nolimits_{q = 1}^{{J_i}} {\exp \left( {\beta '{x_{iq}}} \right)} }} The above equation is the basis for formulating a model for the problem under study. Assuming that employment utility is influenced by work attributes (based on the list of attributes identified in our empirical study: managerial position [SKR], seniority [STZ], age [WIK], lack of tension and stress [BNS], high autonomy [DZS], the opportunity for personal development [MRO], work in accordance with skills [PZU], convenient working hours [DGP], long vacation [DŁU], the level of social respect related to the job [ZPL] and adequate pay [ODP]), the utility of choosing job offer j takes the following form: Uj=β1SKRj+β2STZj+β3WIKj+β4BNSj+β5DZSj++β6MROj+β7PZUj+β8DGPj+β9DŁUj+β10ZPLj++β11ODPj\matrix{{{U_j} = {\beta _1}{SKR_j} + {\beta _2}{STZ_j} + {\beta _3}{WIK_j} + {\beta _4}{BNS_j} + {\beta _5}{DZS_j} + } \hfill \cr { + {\beta _6}{MRO_j} + {\beta _7}{PZU_j} + {\beta _8}{DGP_j} + {\beta _9}{DLU_j} + {\beta _{10}}{ZPL_j} + } \hfill \cr { + {\beta _{11}}{ODP_j}} \hfill \cr } If the assumptions of the MNL model are met, the probability of choosing an alternative j from among the J alternatives can be written as follows: e−(β1SKRj+β2STZj+β3WIKj+β4BNSj+β5DZSj+β6MROj+β7PZUj+β8DGPj+β9DŁUj+β10ZPLj+β11ODPj)∑qj(eβ1SKRj+β2STZj+β3WIKj+β4BNSj+β5DZSj+β6MROj+β7PZUj+β8DGPj+β9DŁUj+β10ZPLj+β11ODPj){{{e^{ - \left( {{\beta _1}{SKR_j} + {\beta _2}{STZ_j} + {\beta _3}{WIK_j} + {\beta _4}{BNS_j} + {\beta _5}{DZS_j} + {\beta _6}{MRO_j} + {\beta _7}{PZU_j} + {\beta _8}{DGP_j} + {\beta _9}D\pounds{U_j} + {\beta _{10}}{ZPL_j} + {\beta _{11}}{ODP_j}} \right)}}} \over {\sum\nolimits_q^j {\left( {{e^{{\beta _1}{SKR_j} + {\beta _2}{STZ_j} + {\beta _3}{WIK_j} + {\beta _4}{BNS_j} + {\beta _5}{DZS_j} + {\beta _6}{MRO_j} + {\beta _7}{PZU_j} + {\beta _8}{DGP_j} + {\beta _9}{DLU_j} + {\beta _{10}}{ZPL_j} + {\beta _{11}}{ODP_j}}}} \right)} }} The MNL model has its limitations. It only reflects differences in preferences regarding the observed characteristics, but it does not account for the variance resulting from unobserved aspects. Based on the MNL model, it is possible to draw conclusions about the dynamics of preferences only in a situation where the unobserved variables are not correlated with time. If the random part of utility is correlated with time, the model has no proper specification. The assumption of the MNL model shows another limitation, saying that utility is independent of other alternatives presented to a respondent. The last objection concerns the results of models of the same class. The results only show the proportion of utility derived from particular attribute levels, making it possible to calculate the marginal rates of substitution (MRS) between attributes, and it is impossible to interpret the parameters without referring them to each other (Train, 2009). In summary, the MNL model is the basic tool for the econometric analysis of choices. However, due to its restrictive assumptions, it does not allow for modelling the heterogeneity of consumers’ preferences, the uncertainty of their choices and irrationality.

Job satisfaction and employee commitment are important aspects of the efficient operation of an organisation. Therefore, increasing numbers of organisations want to ensure high employee morale and undertake research on the satisfaction and commitment of their employees. Job satisfaction and commitment to work are symptoms of complex processes taking place inside an organisation. For this purpose, direct employee satisfaction surveys were conducted with the employees of a selected transport service provider. The following satisfaction factors were used in the study: managerial position, seniority, employee age, lack of tension and stress at work, high autonomy, possibility of personal development, work in accordance with skills, convenient working hours, long vacation, tasks respected by people and adequate pay. The selection of attributes for the empirical study was made on the basis of attributes recognised in the literature. When selecting the attributes, the direct benefits of employment (physical and mental conditions of work performance: lack of tension and stress at work, high autonomy and work in accordance with skills) and indirect, related to the fulfilment of needs in the long term (possibility of personal development, convenient working hours and long vacation). Thus, the employee's level of job satisfaction consisted of utility derived from pay and other forms of remuneration (obtained directly from the employer), as well as utility derived from physical and mental conditions in which work tasks are performed.

Based on the distinguished satisfaction factors, the following RQs were asked: Is employee satisfaction significantly influenced by pay? Do factors other than pay significantly affect the job satisfaction of employees of the transport company? and Does a job position significantly affect employee satisfaction?

The data obtained for the study was based on a survey conducted in one of the companies providing transport services. An online survey was used in the empirical study. Then, the focus was placed on building a questionnaire containing questions that will be submitted to respondents during the study. The first rule regarding the construction of the questionnaire was the selection of questions for the research method. In the case of online research, it was particularly important to clearly formulate questions and use comprehensible language. The questionnaire uses closed-ended questions, i.e. questions with a list of prepared, predefined answers presented to the respondent to choose from. Among the closed questions, alternative questions were used, where the answer possibilities were as follows: ‘definitely satisfactory/definitely yes/yes/always has/rather satisfactory/rather yes/I have no opinion/hard to say/rather unsatisfactory/rather not, and so on’. Most of the questions were disjunctive, i.e. they required choosing only one answer out of more than two answers. There were conjunctive questions that made it possible to choose more than one of the possible answers. In addition, the questionnaire used substantive questions, i.e. questions about the respondent, containing a set of questions about all the demographic and social features of the respondent important for the research, such as employee age, seniority and job position. The questions included in the questionnaire began with the most general question, followed by specific questions. After the questionnaire was prepared, it was sent to the previously selected group of respondents by e-mail with a link that redirected directly to the questionnaire.

The survey was conducted twice. The first survey took place in 2017, when the respondents were asked to assess the quality of working conditions and the satisfaction of employees of the company providing transport services. The next survey was carried out in 2018 after a pay rise by an average gross amount of PLN 250. The main task of the study was to obtain an answer to the question of how employees perceived the company in particular areas of its operation and to identify factors influencing employee satisfaction, such as adequate pay, tasks respected by people, long vacation, convenient working hours, work in accordance with skills, personal development, high autonomy and the lack of tension and stress. The above-mentioned factors are internal factors of the company, so the company is able to directly influence the level of satisfaction. Therefore, the aim of the study was to determine the factor or factors providing satisfaction to employees working in the enterprise providing transport services.

The addressees of the study were administrative employees and lower-level managers of the enterprise providing transport services. The 2017 survey involved 289 people, while 316 employees took part in the 2018 survey.

Initial verification of the data obtained from the study and the RQs posed allowed for the formulation of the following hypotheses:

Hypothesis 1: Employee satisfaction is significantly influenced by pay.

Hypothesis 2: Factors other than pay significantly affect job satisfaction.

Hypothesis 3: The job position held by an employee significantly influences their satisfaction.

The starting point for the study was the use of the MNL model. In this model, the dependent variable is discrete and takes values from a countable and finite set of values with a specific hierarchy.

The level of satisfaction was established based on the respondents’ answers. The sum of the points for the responses from each respondent made it possible to determine the level of satisfaction of an employee of the surveyed company providing transport services.

If the sum of the respondent's answers was <5, the employee was definitely dissatisfied with the work in the company providing transport services (Level 1, y < 5). If the sum of the respondent's answers was <10, but >5, the employee was dissatisfied (Level 2, 5 £ y < 10). If the sum of the respondent's answers was <15 but >10, the employee was satisfied (Level 3, 10 £ y < 15). Finally, if the sum of the respondent's answers was >15, the employee was definitely satisfied with the work in the company providing transport services (Level 4, y ≥ 15).