Determinants of the Low SME Loan Approval Rate in Croatia: A Latent Variables Structural Equations Approach

Name of client
CERGE-IE / Global Development Network of the World Bank
Ministry of Science, Education and Sports, RC

Origin of funding
CERGE-IE / Global Development Network of the World Bank
Ministry of Science, Education and Sports, RC

Dates (start/end)
9/2001 - 5/2002

Name of partners
ECODES-Agriconsulting, Bruxelles, Belgium; Croatian Chamber of Commerce, Zagreb

Description of project - Overall objective:
The aim of the project is to determine the main causes of the low SME loan application approval rate in Croatia, and to analyse structural relationships among them, thus providing scientific evidence needed as policy input in designing and implementing the SME financial policy. Methodologically, we cannot estimate the probability or odds of acceptance of the loan applications given their exogenous properties because consistency in assessment criteria and general banks’ attitudes toward SME lending are among key research questions. These problems can be overcome by the use of structural equations modelling with latent variables by specifying a latent system of equations for the unobserved variables and measurement systems for the observed indicators.

Type of services provided

  • development of a comprehensive data base registry on the SME loan applicants (3,600 entries) coding 12 indicators derived from the loan applications submitted through the “Gruda snijega” programme
  • data collection (in 21 Counties) with CAPI method / data collected from regional County offices and commercial banks
  • data analysis and econometric estimation of latent variable models
  • determining covariance structure of loan application process, comparing regions and banks
  • testing consistency of loan application criteria and comparing criteria across commercial banks
  • presentation of project results in seminar / publishing of the report