Modeling Potential Productivity Gains from SME Growth: A Monte Carlo Simulation for Turkish Manufacturing Firms
Purpose: The paper estimates the labor-productivity gains that Turkish manufacturing small- and medium-sized enterprises (SMEs) can expect when they scale up from micro to small and from small to medium size classes.Methodology: Using value-added-per-employee data from TurkStat, we run Monte Carlo s...
Saved in:
Main Author: | |
---|---|
Format: | Article |
Language: | English |
Published: |
Sanayi ve Teknoloji Bakanlığı
2025-07-01
|
Series: | Verimlilik Dergisi |
Subjects: | |
Online Access: | https://dergipark.org.tr/tr/download/article-file/4824058 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Purpose: The paper estimates the labor-productivity gains that Turkish manufacturing small- and medium-sized enterprises (SMEs) can expect when they scale up from micro to small and from small to medium size classes.Methodology: Using value-added-per-employee data from TurkStat, we run Monte Carlo simulations that keep the technological tier constant while varying firm size. Log-normal and gamma shock processes are combined with three coefficients of variation (0.5, 1.0, 2.0) to represent alternative degrees of within-class heterogeneity.Findings: High-technology SMEs realize the highest average gains—around TL 260–370 thousand—especially in the small-to-medium transition. Low-technology firms show modest mean improvements and a rising probability of negative outcomes as heterogeneity increases. Log-normal shocks generate fatter upper tails and more extreme winners, whereas gamma shocks deliver narrower central ranges but still sizable outliers at high dispersion.Originality: This is among the first studies to quantify potential productivity pay-offs from SME scaling in Turkey within a fully stochastic framework. By modeling technology-conditioned heterogeneity with aggregate data, it offers fresh evidence for designing conditional subsidies and other targeted industrial-policy tools. |
---|---|
ISSN: | 1013-1388 |