Technological Capabilities and Learning in African Enterprises

Abstract

This study is one of the first systematic attempts to assess technological capabilities of African manufacturing firms, to estimate their technical efficiency levels and to analyze the significance of technological capabilities in determining firm level technical efficiency. Technological capabilities are broadly defined as the skills and knowledge needed to set up and efficiently operate modern industry. Using firm level data in Ghana, Kenya and Zimbabwe, the study evaluates components of investment capabilities, production capabilities and learning capabilities. Four industries provide the focus in each country: textiles and garments, wood working, metal working and food processing. It is shown that there are important cross-country differences in technological capabilities of firms, as well as across the four sectors of manufacturing activity. At the same time, although significant private efforts are being made by entrepreneurs in each country to upgrade their technological capabilities, the study finds that overall technological capabilities are low in these economies relative to the rest of the world. Econometric analysis using stochastic frontier methods is then undertaken to estimate levels of technical efficiency is a distinguishing aspect of the study, marking the first such comprehensive attempt across sub-Saharan economies. Analysis of technical efficiency shows that technological capabilities are significant determinants of efficiency at the firm, sector and country levels. The study also shows that the traditional learning and linkage mechanisms by which firms enhance their technological competence are weak and, many times, missing in African countries. Recommendations are offered for how development agencies can assist in building up these learning mechanisms, both in their private and collective forms.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 148 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.