Working Paper: CEPR ID: DP9758
Authors: Olivier Cadot; Ana M. Fernandes; Julien Gourdon; Aaditya Mattoo
Abstract: This paper evaluates the effects of the FAMEX export promotion program in Tunisia on the performance of beneficiary firms. While most studies assess only the short-term impact of such programs, we consider also the longer-term impact. Estimates suggest that beneficiaries initially saw both faster export growth and greater diversification across destinations and products. However, three years after the intervention, beneficiaries? growth rates and export levels were not significantly different from those of a control group even though they remained more diversified. We confirm that this divergence between export growth and diversification is not due to small export transactions to new markets creating an illusion of diversification; to greater exposure of beneficiary firms to crisis-affected economies leading to stunted export growth; or to spillover benefits for non-beneficiary firms resulting in their catching-up in export sales. We find some evidence that the divergence may be related to constraints within the firm, such as limited experience and in-house export capacity; to external constraints, such as access to finance; and to the design and implementation of the FAMEX program which placed greater emphasis on diversification than on export growth.
Keywords: export margins; export promotion; firms; impact evaluation; matching grant; propensity-score matching; Tunisia
JEL Codes: C23; F13; F14; L15; L25; O17; O24
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| export growth differences diminish over time (F62) | internal constraints and external factors (O36) |
| Famex program (C88) | export outcomes for beneficiary firms (F10) |
| Famex program (C88) | faster export growth (F10) |
| Famex program (C88) | greater diversification across destinations and products (Z39) |
| faster export growth (F10) | diversification (G11) |
| export levels of beneficiaries (H55) | not significantly different from control group (C92) |