This thesis analyses the productivity of Local Public Transport (LPT) companies in Italy, with the aim of distinguishing the performance component attributable to firm manage ment from the socio-economic factors that characterise the operating context. The analysis starts with the construction of a database based on annual reports and financial statements of 50 Italian operators with over 100 employees, referring to the year 2023. An initial descriptive analysis is conducted on this sample in order to document the high structural heterogeneity of the sector at national level. Based on this evidence, an econometric analysis is developed using a two-stage Partial Factor Productivity (PFP) model. This phase uses a restricted panel of 15 operators observed over the three-year period 2021–2023, selected on the basis of the complete availability of financial, operational and contextual data. In the first stage, gross productivity is estimated using a multilateral Translog index, which aggregates inputs and outputs through cost/revenue weights and returns a comparable measure of the input-output ratio between operators. In the second stage, a log-linear regression model is estimated to explain gross productivity as a function of structural socio-economic variables. The regression residuals are interpreted as a measure of net productivity, adjusted for the influence of contextual factors. The results show that the unemployment rate, motorisation rate, modal share of public transport, income per capita and modal configuration of the service explain a significant proportion of the observed variability. This suggests that the differences in productivity observed between operators cannot be interpreted solely as an expression of managerial performance, but statistically reflect the structural and socio-economic conditions of the operating environment, with important implications for benchmarking activities and public resource allocation mechanisms.
La presente tesi analizza la produttività delle imprese di Trasporto Pubblico Locale (TPL) in Italia, con l’obiettivo di distinguere la componente di performance imputabile alla gestione aziendale dai fattori socio-economici che caratterizzano il contesto operativo. L’analisi parte dalla costruzione di un database basato su relazioni annuali e bilanci aziendali di 50 operatori italiani con oltre 100 dipendenti, riferiti all’anno 2023. Su questo campione viene condotta una prima fase di analisi descrittiva finalizzata a documentare l’elevata eterogeneità strutturale del settore a livello nazionale. Sulla base di tali evidenze, viene sviluppata un’analisi econometrica attraverso un modello two-stage Partial Factor Productivity (PFP). Questa fase utilizza un panel ristretto di 15 operatori osservati nel triennio 2021–2023, selezionati in funzione della disponibilità completa di dati finanziari, operativi e contestuali. Nel primo stadio viene stimata la produttività lorda mediante un indice multilateral Translog, che aggrega input e output attraverso pesi di costo/ricavo e restituisce una misura comparabile del rapporto input-output tra operatori. Nel secondo stadio, viene stimato un modello di regressione log-lineare in cui la produttività lorda è spiegata da variabili socio-economiche strutturali. I residui della regressione sono interpretati come misura di produttività netta, corretta per l’influenza dei fattori di contesto. I risultati mostrano che il tasso di disoccupazione, tasso di motorizzazione, quota modale del trasporto pubblico, reddito pro capite e configurazione modale del servizio spiegano una quota rilevante della variabilità osservata. Questo suggerisce che le differenze di produttività osservate tra operatori non possono essere interpretate esclusivamente come espressione di performance gestionale, ma riflettano statisticamente le condizioni strutturali e socio-economiche dell’ambiente operativo, con rilevanti implicazioni per le attività di benchmarking e per i meccanismi di allocazione delle risorse pubbliche.
Assessing productivity of local public transport companies
Cabras, Lorenzo
2024/2025
Abstract
This thesis analyses the productivity of Local Public Transport (LPT) companies in Italy, with the aim of distinguishing the performance component attributable to firm manage ment from the socio-economic factors that characterise the operating context. The analysis starts with the construction of a database based on annual reports and financial statements of 50 Italian operators with over 100 employees, referring to the year 2023. An initial descriptive analysis is conducted on this sample in order to document the high structural heterogeneity of the sector at national level. Based on this evidence, an econometric analysis is developed using a two-stage Partial Factor Productivity (PFP) model. This phase uses a restricted panel of 15 operators observed over the three-year period 2021–2023, selected on the basis of the complete availability of financial, operational and contextual data. In the first stage, gross productivity is estimated using a multilateral Translog index, which aggregates inputs and outputs through cost/revenue weights and returns a comparable measure of the input-output ratio between operators. In the second stage, a log-linear regression model is estimated to explain gross productivity as a function of structural socio-economic variables. The regression residuals are interpreted as a measure of net productivity, adjusted for the influence of contextual factors. The results show that the unemployment rate, motorisation rate, modal share of public transport, income per capita and modal configuration of the service explain a significant proportion of the observed variability. This suggests that the differences in productivity observed between operators cannot be interpreted solely as an expression of managerial performance, but statistically reflect the structural and socio-economic conditions of the operating environment, with important implications for benchmarking activities and public resource allocation mechanisms.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_03_Cabras_Tesi.pdf
non accessibile
Descrizione: Tesi
Dimensione
2.99 MB
Formato
Adobe PDF
|
2.99 MB | Adobe PDF | Visualizza/Apri |
|
2026_03_Cabras_Executive_Summary.pdf
non accessibile
Descrizione: Executive Summary
Dimensione
952.23 kB
Formato
Adobe PDF
|
952.23 kB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/253116