Estimation of Firm Labour Productivity and Sales Growth from Artificial Intelligence in Sub-Saharan African Countries

Olanrewaju Adewole Adediran

doi:10.12688/f1000research.179912.1

Home Browse Estimation of Firm Labour Productivity and Sales Growth from Artificial...

ALL Metrics

-

Views

-

Downloads

Get PDF

Get XML

Export

▬

✚

Research Article

Estimation of Firm Labour Productivity and Sales Growth from Artificial Intelligence in Sub-Saharan African Countries

[version 1; peer review: 1 approved with reservations]

Olanrewaju Adewole Adediran

PUBLISHED 12 May 2026

Author details Author details

Sustainable Livelihoods, University of South Africa Graduate School of Business Leadership, Johannesburg, Gauteng, 1686, South Africa

Olanrewaju Adewole Adediran
Roles: Conceptualization, Data Curation, Formal Analysis, Investigation, Methodology, Software, Supervision, Validation, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing

OPEN PEER REVIEW

REVIEWER STATUS

This article is included in the Artificial Intelligence and Machine Learning gateway.

Abstract

The fast integration of artificial intelligence (AI) into business operations and labour processes is reshaping global economic landscapes. The study examines the effects of AI adoption on labour productivity and sales growth in selected sub-Saharan African countries using a firm-level dataset from the World Bank Enterprises from 2007 to 2024. The study employs feasible generalised least squares (FGLS), robust ordinary least squares (OLS), and high-dimensional fixed effects (HDFE) linear regression techniques. The results show that AI has a significant positive relationship with firm labour productivity and sales growth in the selected sub-Saharan African countries. However, nuances differ across countries due to varying industrial structures. Results vary across the 10 selected countries due to differences in technological readiness. These results underscore the importance of targeted policy interventions, such as upskilling initiatives and supportive regulatory frameworks, to harness AI’s benefits while mitigating adverse impacts on workers. This research contributes to the growing body of literature on technology adoption in developing economies, offering policymakers and business leaders in sub-Saharan Africa valuable insights.

Keywords

Artificial intelligence, labour productivity growth, sales growth, high-dimensional fixed effects (HDFE) linear regression

Corresponding author: Olanrewaju Adewole Adediran

Competing interests: No competing interests were disclosed.

Grant information: The author(s) declared that no grants were involved in supporting this work.

Copyright: © 2026 Adewole Adediran O. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite: Adewole Adediran O. Estimation of Firm Labour Productivity and Sales Growth from Artificial Intelligence in Sub-Saharan African Countries [version 1; peer review: 1 approved with reservations]. F1000Research 2026, 15:711 (https://doi.org/10.12688/f1000research.179912.1) First published: 12 May 2026, 15:711 (https://doi.org/10.12688/f1000research.179912.1) Latest published: 12 May 2026, 15:711 (https://doi.org/10.12688/f1000research.179912.1)

1. Introduction

Artificial intelligence (AI) refers to the development, application, and implementation of new technology (or computer programmes) to solve or simplify tasks that require human ability,^1–3 as well as to produce new goods and services or improve their production. It is challenging to define AI precisely. Existing studies have revealed that AI could be influenced by factors such as organisational administration, government policy,^4–7 globalisation, among others. Despite the enormous benefit of adopting AI, studies have argued that it could create skill mismatches and labour displacement,^8,9 which might affect productivity and sales.^10,11 Other limitations to the adoption of AI may include limited infrastructure, a lack of know-how or expertise, limited access to affordable AI technology, poor data ecosystems, and a lack of government policy that supports AI,^12,13 especially in sub-Saharan African countries.

AI is known as a catalyst for increased firm labour productivity.^14,15 However, asymmetric information within businesses^16,17 may limit the adoption of AI. In such cases, the firm’s productivity slows down for firms without AI technologies and speeds up for firms with technological expertise. As the AI adoption may improve labour productivity, it could also enhance sales. AI adoption provides a digital platform for competitive advantage, influencing marketing, increasing customer awareness, and driving sales.^18,19 Despite increasing adoption of artificial intelligence among firms, its actual impact on labour productivity and sales growth remains ambiguous, particularly in emerging economies where digital infrastructure, workforce skills, and organisational capacity differ significantly. The study provides answers to the following questions: Does AI adoption increase firm labour productivity and translate into higher sales growth? Are AI effects heterogeneous across firm size, sector, and countries?

The theoretical understanding of AI and labour outcomes relied on neoclassical growth theory. The growth theory started with “The Wealth of Nations”, which was first postulated by Adam Smith in the 1950s.²⁰ When there is steady, constant growth, national savings are likely to equal the capital-output ratio and the growth rate of the active labour force (or workers). However, as the capital-output ratio (which implies technology, including AI) increases, there is a tendency for the employment of skilled workers (professionals or those in information and communication technology (ICT)) to rise, potentially leaving unskilled workers unemployed. The Neoclassical school of thought holds that the introduction of technology (changes in technology and increases in aggregate production) is likely to contribute to steady growth. On the other hand, improvements in the production of goods and services (driven by technologies such as AI) are likely to increase national sales and improve life satisfaction.

Empirically, existing studies have found a positive association between AI and labour productivity.^2,13,21 While studies have shown that AI may affect labour productivity, it may also lead to a decline in labour income due to the uniform displacement of repetitive tasks.³ Also, the introduction of Artificial Intelligence (AI) into business activities is likely to drive sales growth. AI mechanisms enhance operational efficiency and sales strategies.^22,23 AI tends to influence the dynamics of marketing capacity.^24–26 Firms are relying on AI to plan sales strategies by understanding customer choice, conducting internet marketing, forecasting,^27,28 and increasing turnover. Despite the enormous benefits of AI for sales growth, there are hindrances to its adoption, including the high cost of access and a lack of manpower and technology know-how.^29,30

In addition, a strand of the literature uses primary data collected from 697 people in small and medium enterprises, and a reliability test was conducted on the data.³¹ Factor analysis was employed to construct an index of key variables used. It employed the SEM technique to investigate the relationships among AI adoption, workforce management, and marketing and sales. The study revealed that AI helps in labour force and business HR decision-making capacity. AI has a significant positive relationship with business and marketing. However, when small and medium enterprises have limited resources and innovative initiative, it would lead to a lack of technical knowledge and skilled personnel,³¹ especially when operating in areas where institutions are unwilling to improve. Pointing out that small and medium enterprises have challenges that may prevent them from adopting AI,³² such as a lack of a digital framework.

It is interesting to note that there is a paucity of empirical studies on how AI enhance labour productivity²¹ and sales growth.³³ As mentioned earlier, the study was based on neoclassical theory and hypothesised that AI has the potential to improve labour productivity and sales growth. Additionally, the current study employed a high-dimensional fixed effects (HDFE) estimator to address a unique question: whether AI adoption is likely to influence labour productivity growth in large labour-intensive economies and sales growth, with evidence from selected countries in sub-Saharan Africa. The remaining sections of the study include Section 2, which presents the data and methods; Section 3, which presents the empirical results; and Section 4, which concludes the study.

2. Data and methods

The current study utilises a dataset from the World Bank’s Enterprise Surveys, covering various countries and years from 2007 to 2024. The World Bank Enterprise Surveys (WBES) are conducted across many countries and provide diverse firm-level data on the business environment, including conditions that affect firms. The data underlying this study are publicly available from the original data provider, the World Bank Enterprise Surveys database, at.⁴⁷ The study uses a panel dataset from 10 selected countries, including the Central African Republic, Ethiopia, Ghana, Kenya, Lesotho, Nigeria, Rwanda, Sierra Leone, South Africa, and Tanzania, covering 2007 to 2024. Variables were sorted for each country before setting a data panel. It is important to note that the question collects both subjective and objective data on firm business conditions.^34,35 The survey uses standardised questionnaires and uniform sampling methods. They produce rich data concerning a firm’s characteristics, as well as information on perceived constraints, including labour productivity, sales growth, innovation-related questions,³⁶ firm age, and international certification, among others. Data collection was initiated in 2006 in a standardised manner, allowing consistent overtime and comparative analysis across countries.^37,38 Access the dataset used for the analysis via.⁴⁷

The outcome variables are labour productivity growth and sales growth. The study computes labour productivity and sales growth following the methods of.^2,39 First, the labour productivity growth variable was generated by subtracting the number of permanent full-time workers three financial years ago (labour 3 years ago) from the number of permanent full-time workers at the end of the last financial year (size of employment) as reported in the survey questionnaire. The study generated the previous labour value by dividing past sales dollar conversion by labour in 3 years ago, and the current labour value by dividing current sales dollar conversion by current labour full-time or the size of employment. The study calculates labour productivity growth by generating the current labour value divided by the previous labour value, multiplied by one divided by the current year minus the previous year, minus one. Sales growth is computed the same way as labour productivity growth. The sales-related question is: what are this establishment’s total annual sales for all products and services (current sales), and what were its sales three years ago (previous sales). The study generated sales-dollar conversion for both current and prior sales. Sales growth was calculated by dividing the current year’s sales by the previous year’s sales, minus the difference between the two years.

The current study is unique in that it focuses on selected African countries, rather than a cross-sectional study like,³⁹ which examines the association between innovation and firm performance in India. Uses WBES 2013–14 and the probit model. The study found that long years spent in informal and unregistered firms lead to faster innovation than in other firms.² employed a dataset from the Community Innovation Survey (CIS) 2018–2019 and the 2SLS technique (with an instrumental variable, IV) to investigate AI adoption and firm productivity in Germany. The study remarks that AI improves firm productivity.

Furthermore, the AI was derived by generating an index of various innovation questions, which comprises New or significantly improved products introduced in the last three years. An improved organisational structure has been introduced in the last 3 years. Over the last three years, has it provided employees with sufficient time to develop new ideas, and has it allocated funds for formal research and development activities? The study uses multiple correspondence analyses to generate the AI index. Additionally, the study creates information asymmetry through mechanisms of international certification, including a firm being Publicly Listed on a stock exchange (with or without shareholders) and undergoing an external audit. It is interesting to note that variables for artificial intelligence and information asymmetry are dummy variables. In the computation, the study uses the MCA command in Stata version 15 to predict the index, then multiplies it by −1 to account for possible errors in the coordinate. Other variables include firm size (micro, small, medium, and large firms), firm age, and industry (a dummy variable).

Table 1 presents a summary of the dependent variable, labour productivity growth, with a mean of 0.102 and a range of −0.927 to 195.32. Sales growth ranges from −1 to 36. The average age of the firm is 17 years, ranging from 0 to 220 years. The average of the IA index is −0.04, and the standard deviation is approximately 1. It ranges between −1.62 and 5.12. Generally, the standard deviation measures the dispersion of a dataset from its mean (average), indicating that the data is likely to be consistent for all stages of the analysis.

Table 1. Descriptive Statistics.

Variable	Obs	Mean	Std. Dev.	Min	Max
Labour Productivity Growth	11837	0.102	2.731	−0.927	195.32
AI index	12578	−0.089	0.807	−1.441	7.087
Firm age	12578	17.014	14.397	0	220
Firm size
Micro	11968	0.153	0.36	0	1
Small	11968	0.53	0.499	0	1
Medium	11968	0.235	0.424	0	1
Large	11968	0.082	0.275	0	1
IA index	12579	−0.039	0.951	−1.622	5.124
Sales growth	11889	−0.048	0.998	−1	36.283
Foreign own	12579	0.071	0.257	0	1
Industries
Fabricated metal	8842	0.038	0.191	0	1
Food and beverages	11749	0.086	0.281	0	1
Furniture	7034	0.015	0.12	0	1
Garments	10822	0.055	0.228	0	1
Nonmetallic	9014	0.032	0.176	0	1
Publishing	7034	0.015	0.12	0	1
Other manufacturing	12579	0.109	0.312	0	1
Other services	12579	0.103	0.304	0	1
Repair sales of motor vehicles	7034	0.011	0.102	0	1
Wholesales	7034	0.016	0.127	0	1
Hotel and restaurant	10019	0.043	0.204	0	1
Retails	12049	0.09	0.287	0	1
Transportation	8842	0.025	0.156	0	1
Manufacturing panel	8842	0.036	0.187	0	1
Retail panel	7034	0.023	0.149	0	1
Other service panel	7034	0.018	0.132	0	1

2.1 Methods

The study first employed robust ordinary least squares (OLS), which examines the relationship between the dependent and independent variables and controls for model stability and estimation despite uncertainty, disturbances, or outliers. Additionally, the employed high-dimensional fixed-effect estimator accounts for unobserved heterogeneity that may otherwise bias the results and is nested within a robust standard error, rather than absorbing cluster identification.

The model is as follows;

Labour {productvity growth}_{it} = β_{0} + β_{1} {AI}_{it} + β_{2} Firm {age}_{it} + β_{3} {IA}_{it} + β_{4} X_{it} + ε_{it}

Where ${AI}_{it}$ denote artificial intelligence; ${IA}_{it}$ implies information asymmetry; $X_{it}$ denote other independent variables (such as foreign ownership; legal status included in country-specific estimation). $β_{0}$ is an intercept or a constant. $β_{1}$ , $β_{2}, β_{3},$ and $β_{4}$ are the coefficients of the regressor and control variables.

3. Results and interpretation

Table 2 presents a correlation between the dependent and independent variables. The correlation suggests that labour production growth and the AI index are likely to be positively correlated (r = 0.019). Labour productivity and information asymmetry are positively correlated (r = 0.031). Labour productivity and foreign ownership are positively correlated (r = 0.018). Labour productivity and firm age are negatively correlated (r = −0.020). Existing studies have found that a firm’s age is negatively associated with productivity.^40,41

Table 2. Matrix of correlations.

Variables	(1)	(2)	(3)	(4)	(5)
(1) Labour Productivity G	1.000
(2) AI	0.019	1.000
(3) Firm age	−0.020	0.080	1.000
(4) IA index	0.031	0.245	0.187	1.000
(5) Foreign-owned	0.018	0.124	0.115	0.238	1.000

Table 3 presents the VIFs of the independent variables, which are moderately correlated as they satisfy the rule of thumb. The rule of thumb for the variance inflation factor (VIF) states that VIFs should be less than 10. If it is less than 10, it suggests the possibility of multicollinearity in the analysis.

Table 3. Variance inflation factor.

	VIF	1/VIF
IA index	1.143	0.875
foreign own	1.071	0.934
AI index	1.07	0.935
Firm age	1.043	0.959
Mean VIF	1.082

In Table 4 (1), the study employs feasible generalised least squares (FGLS) to control for autocorrelation in a model with no precise error structure. The study performed ordinary least squares (OLS) regression, obtained residuals, estimated the variance, assessed model heteroscedasticity, predicted the variance, generated weights, and plugged them into the model. Other columns (2–5) applied OLS robust. Table 4 column (1) presents that there is a significant positive relationship between labour productivity growth and AI.² Also, information asymmetry has a significant positive relationship with labour productivity growth. However, firm age and foreign ownership of business have a significant negative effect on labour productivity growth. Column (2) presents that the information asymmetry has a significant positive relationship with labour productivity growth for micro-scale enterprises. Sales growth has a significant positive relationship with labour productivity. However, firm age has a significant negative relationship with labour productivity growth. This implies that, as firms age, they tend to be unable to keep pace with technological advancements in sub-Saharan Africa. There is a positive relationship between labour productivity and sales growth. Sales growth in micro-scale enterprises is higher than in other types of enterprises, such as small-scale enterprises.

Table 4. Labour productivity growth and AI: robust OLS.

	(1)	(2)	(3)	(4)	(5)
	Panel	Micro	Small	Medium	Large
Variables	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth
AI	0.00702*	−0.0163	0.0235	−0.0170	0.0261
AI	(0.00427)	(0.0784)	(0.0285)	(0.0208)	(0.0280)
Firm age	−0.000557***	−0.00757	0.000285	−0.00323***	−0.00173***
Firm age	(0.000198)	(0.00486)	(0.00147)	(0.000973)	(0.000390)
IA index	0.00888***	0.416***	0.0166	−0.00339	0.0203
IA index	(0.00332)	(0.149)	(0.0110)	(0.0150)	(0.0124)
Foreign ownership	−0.0242*
Foreign ownership	(0.0146)
Sales Growth		6.265***	1.145***	0.803***	0.790***
Sales Growth		(0.891)	(0.207)	(0.0325)	(0.0661)
Constant	−0.0581***	2.931***	0.0726***	0.126***	0.138***
Constant	(0.00713)	(0.424)	(0.0227)	(0.0405)	(0.0258)
Year dummy	Yes
Observations	11,837	1,737	6,311	2,805	984
R-squared	0.253	0.732	0.430	0.385	0.716

*** p < 0.01,

** p < 0.05,

* p < 0.1; IA index denotes information asymmetry index; AI implies artificial intelligence; Source: computed by the Author 2026

Column (3) presents that the firm age has a significantly negative relationship with labour productivity growth for small-scale enterprises. Sales growth has a significant positive relationship with labour productivity. This is consistent with the literature.⁴² A previous study remarked that innovation (AI technology) is likely to enhance productivity, and an increase in sales growth is likely to increase labour productivity.⁴² Sales growth in small-scale enterprises follows that of micro-scale enterprises. It is higher than medium and large enterprises. This indicates that micro and small enterprises are growing faster than medium and large enterprises in sales and labour productivity.

Column (4) presents that information asymmetry has an insignificant negative relationship with labour productivity growth for medium-scale enterprises. Sales growth has a significant positive relationship with labour productivity. While firm age has a significant negative relationship with labour productivity growth. Column (5) presents that artificial intelligence has an insignificant positive relationship with labour productivity growth for large-scale enterprises. This is expected because large-scale enterprises may be able to expand their production capacity using capital-intensive technology. Also, sales growth has a significant positive relationship with labour productivity. Sales growth in large-scale enterprises is the lowest among all types, including micro, small, and medium enterprises.

Table 5 column (1) presents that there is a significant positive relationship between AI and sales growth.² However, firm age has a significant negative relationship with sales growth. Column (2) presents that AI has an insignificant negative relationship with sales growth for micro-scale enterprises. Column (3) presents that the information asymmetry has a significant positive relationship with sales growth for small-scale enterprises. The firm age has a significant negative relationship with sales growth among micro-scale enterprises. Column (4) presents that firm age has a significant negative relationship with sales growth for medium-scale enterprises. Column (5) presents that firm age has a significant negative relationship with sales growth for large-scale enterprises.

Table 5. Sales growth and AI: robust OLS.

	(1)	(2)	(3)	(4)	(5)
	Panel	Micro	Small	Medium	Large
Variables	Sales Growth	Sales Growth	Sales Growth	Sales Growth	Sales Growth
AI	0.0420***	−0.0101	0.0769***	−0.0290	0.0323
AI	(0.0125)	(0.0278)	(0.0213)	(0.0268)	(0.0331)
Firm age	−0.00116**	0.000292	−0.00385***	−0.00271***	0.00138*
Firm age	(0.000470)	(0.00170)	(0.000689)	(0.000907)	(0.000752)
IA index	0.0172	0.0624	−0.00120	−0.00240	0.0938
IA index	(0.0151)	(0.0434)	(0.0190)	(0.0209)	(0.0593)
Foreign ownership	0.0717
Foreign ownership	(0.0503)
Constant	−0.0150*	−0.363***	0.0991***	0.0567**	−0.216***
Constant	(0.00784)	(0.0510)	(0.0201)	(0.0285)	(0.0473)
Year dummy	Yes
Observations	11,889	1,761	6,332	2,810	986
R-squared	0.141	0.002	0.006	0.003	0.014

*** p < 0.01,

** p < 0.05,

* p < 0.1; IA index denotes information asymmetric index; AI implies artificial intelligence; Source: computed by the Author 2026

High-dimensional fixed effects (HDFE) Linear regression.

To conduct a robustness check of robust OLS, the study employs high-dimensional fixed effects (HDFE) linear regression, with standard errors corrected following.⁴³

Table 6 column (1) presents that AI has an insignificant positive impact on labour productivity growth. The information asymmetry has a significant positive impact on labour productivity growth. While firm age has a significant negative effect on labour productivity growth. Column (2) presents that information asymmetry has a significant positive impact on labour productivity growth for micro-scale enterprises. Sales growth has a significant positive impact on labour productivity. However, the AI has a significant negative impact on labour productivity growth for micro-business enterprises. Column (3) presents that the information asymmetry has an insignificant positive impact on labour productivity growth for small-scale enterprises. Sales growth has a significant positive impact on labour productivity. Column (4) presents that firm age has a significant negative effect on labour productivity growth for medium-scale enterprises. Sales growth has a significant positive impact on labour productivity. Column (5) presents that sales growth has a significant positive impact on labour productivity for large-scale enterprises.

Table 6. Labour productivity and AI using the HDFE estimator.

	(1)	(2)	(3)	(4)	(5)
	Panel	Micro	Small	Medium	Large
Variables	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth
AI	0.0245	−0.151*	−0.0140	−0.0323	0.00689
AI	(0.0316)	(0.0804)	(0.0205)	(0.0210)	(0.0196)
Firm age	−0.00352*	0.00296	1.41e-05	−0.00216*	−0.00111
Firm age	(0.00181)	(0.00571)	(0.00126)	(0.00122)	(0.000846)
IA index	0.0892***	0.403***	0.00809	−0.00502	0.0126
IA index	(0.0280)	(0.0838)	(0.0206)	(0.0172)	(0.0166)
Foreign ownership	0.0282
Foreign ownership	(0.0989)
Sales growth		6.626***	1.246***	0.851***	0.837***
Sales growth	(0.0838)	(0.0176)	(0.0203)	(0.0162)
Constant	0.164***	2.878***	0.0670***	0.106***	0.136***
Constant	(0.0407)	(0.120)	(0.0246)	(0.0301)	(0.0311)
Observations	11,837	1,737	6,311	2,805	984
R-squared	0.020	0.790	0.466	0.426	0.755

*** p < 0.01,

** p < 0.05,

* p < 0.1; IA index denotes information asymmetric index; AI implies artificial intelligence; Source: computed by the Author 2026

Table 7 column (1) presents that AI has a significant positive impact on sales growth. Information asymmetry has a significant positive impact on sales growth. Foreign ownership has a significant positive impact on sales growth. However, firm age has a significant negative effect on sales growth. It is worth noting that labour efficiency is likely to lead to higher output⁴² and increased sales. Also, when a firm adopts AI, the production process is likely to be enhanced, leading to improvement in products and better sales performance.

Table 7. Sales growth and AI using the HDFE estimator.

	(1)	(2)	(3)	(4)	(5)
		Micro	Small	Medium	Large
Variables	Sales Growth	Sales Growth	Sales Growth	Sales Growth	Sales Growth
AI	0.0420***	−0.00694	0.0916***	−0.0124	0.0684*
AI	(0.0108)	(0.0335)	(0.0146)	(0.0196)	(0.0388)
Firm age	−0.00116*	−0.000616	−0.00203**	−0.000363	0.000232
Firm age	(0.000620)	(0.00238)	(0.000902)	(0.00114)	(0.00168)
IA index	0.0172*	0.0579*	0.0172	−0.0150	0.0544*
IA index	(0.00956)	(0.0348)	(0.0147)	(0.0160)	(0.0327)
Foreign ownership	0.0717**
Foreign ownership	(0.0337)
Constant	−0.0301**	−0.350***	0.0766***	0.0118	−0.165***
Constant	(0.0139)	(0.0485)	(0.0176)	(0.0281)	(0.0613)
Observations	11,889	1,761	6,332	2,810	986
R-squared	0.141	0.029	0.166	0.153	0.153

*** p < 0.01,

** p < 0.05,

* p < 0.1; IA index denotes information asymmetric index; AI implies artificial intelligence;

Column (2) presents that information asymmetry has a significant positive impact on sales growth for micro-scale enterprises. Column (3) presents that AI has a significantly positive influence on sales growth for small-scale enterprises. Firm age has a significant negative impact on sales growth for small-scale enterprises. Column (4) presents that AI has an insignificant negative impact on sales growth for medium-scale enterprises. Column (5) presents that artificial intelligence has a significant positive effect on sales growth for large-scale enterprises. Also, information asymmetry has a significant positive impact on sales growth. It is plausible that small enterprises experience higher sales growth than large enterprises because they are better able to adapt to market changes and rely less on management decision-making. Large firms are likely to have administrative procedures that may slow decision-making in the marketplace. Micro and small enterprises have faster decision-making processes.

Table 8 column (1) presents that sales growth has a significant negative relationship with labour productivity growth for the Central African Republic (CAR). Having a legal status did not increase the labour productivity growth in CAR. Notwithstanding, IA IA index denotes information asymmetric index; AI implies artificial intelligence; Legal status means company with shareholders and without shareholders Source: computed by the Author 2026 is likely to increase labour productivity growth in the CAR. Column (2) presents that sales growth and firm age have a significant positive relationship with labour productivity growth for Ethiopia. However, AI is unlikely to increase labour productivity in Ethiopia. Column (3) displays that sales growth has a significant positive relationship with labour productivity growth for Ghana. Column (4) presents that the AI has a significant negative relationship with labour productivity growth for Kenya. Firm age and sales growth are positively and significantly related to labour productivity within the same country. Column (5) presents that artificial intelligence has a significant positive relationship with labour productivity growth for Lesotho. While sales growth has a negative relationship with labour productivity growth. The reasons for negative sales growth in Lesotho may include institutional factors, firm-level challenges, infrastructural constraints, and external market competition, among others. Neighbouring countries such as South Africa may be strong competitors to Lesotho in the external market. Hence, their sales may eventually decline.

Table 8. Labour productivity and AI by country.

	(1)	(2)	(3)	(4)	(5)
	Central African Republic	Ethiopia	Ghana	Kenya	Lesotho
Variables	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth
AI	−7.04e-07	−0.00981***	−0.0883	−0.00189***	4.36e-05**
AI	(3.64e-05)	(0.00271)	(0.111)	(0.000580)	(2.03e-05)
Firm age	−5.61e-06	0.000955***	0.0341	0.000199***	−8.16e-07
Firm age	(4.01e-06)	(0.000177)	(0.0242)	(3.28e-05)	(1.60e-06)
IA index	0.00106**	0.00431	0.411	0.00125	−3.75e-05
IA index	(0.000512)	(0.00341)	(0.300)	(0.000846)	(2.28e-05)
Legal status	−0.000261*	−0.000139	−0.463	0.00161	7.37e-05
Legal status	(0.000149)	(0.00841)	(0.484)	(0.00182)	(6.54e-05)
Sales growth	−0.992***	1.718***	7.555***	0.903***	−0.937***
Sales growth	(0.0629)	(0.0605)	(0.332)	(0.0143)	(0.0754)
Constant	0.000613***	0.789***	6.598***	−0.0474***	2.81e-05
Constant	(0.000230)	(0.0371)	(0.279)	(0.00518)	(4.22e-05)
Observations	78	901	479	1,901	124
R-squared	0.749	0.755	0.944	0.854	0.823

*** p < 0.01,

** p < 0.05,

* p < 0.1; IA index denotes information asymmetric index; AI implies artificial intelligence; Legal status means company with shareholders and without shareholders Source: computed by the Author 2026

Table 9 column (1) presents that AI has a significant positive relationship with labour productivity growth in Nigeria. Similarly, sales growth has a significant positive relationship with labour productivity growth. Column (2) presents that sales growth and firm age have a significant positive relationship with labour productivity growth for Rwanda. Column (3) displays that sales growth has a significant positive relationship with labour productivity growth for Sierra Leone. Informatin asymmetry may likely reduce labour productivity growth in Sierra Leone. Column (4) presents that AI has a significant positive relationship with labour productivity growth in South Africa. Also, firm age has a significant positive relationship with labour productivity growth in South Africa. Similarly, sales growth has a significant positive relationship with labour productivity growth. Legal status (shareholder involvement) may likely reduce labour productivity growth in South Africa. Column (5) presents that firm age has a significant negative relationship with labour productivity growth for Tanzania. Also, legal status and sales growth have a significant negative relationship with labour productivity growth.

Table 9. Labour productivity and AI by country.

	(1)	(2)	(3)	(4)	(5)
	Nigeria	Rwanda	Sierra Leone	South Africa	Tanzania
Variables	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth
AI	0.0146***	−0.0955	0.000627	0.00194***	1.20e-05
AI	(0.00558)	(0.104)	(0.00196)	(0.000526)	(8.36e-06)
Firm age	0.000294	−0.00669	5.91e-05	0.000113***	−1.01e-06*
Firm age	(0.000239)	(0.00412)	(0.000218)	(2.20e-05)	(5.17e-07)
IA index	−0.00326	−0.00259	−0.00479**	−0.000616	4.33e-06
IA index	(0.00438)	(0.138)	(0.00219)	(0.000596)	(9.62e-06)
Legal status	0.0268	0.0542	0.0142	−0.00443*	−8.04e-05*
Legal status	(0.0325)	(0.337)	(0.0137)	(0.00251)	(4.62e-05)
Sales growth	0.935***	529.8***	0.904***	0.862***	−0.925***
Sales growth	(0.0164)	(176.4)	(0.0473)	(0.0237)	(0.0345)
Constant	−0.0329***	529.5***	−0.0207***	−0.0186***	8.97e-05***
Constant	(0.00525)	(176.2)	(0.00582)	(0.00142)	(1.49e-05)
Observations	5,500	299	326	1,807	422
R-squared	0.981	0.390	0.628	0.786	0.738

*** p < 0.01,

** p < 0.05,

* p < 0.1; IA index denotes information asymmetric index; AI implies artificial intelligence; Legal status means company with shareholders and without shareholders Source: computed by the Author 2026

Table 10 column (1) presents that AI has an insignificant negative relationship with sales growth for the Central African Republic. Column (2) presents that AI has a significant positive relationship with sales growth for Ethiopia. Shareholders’ involvement is likely to increase sales growth in Ethiopia. Firm age has a significant negative relationship with labour productivity. It is surprising that IA may reduce sales growth in Ethiopia. Column (3) displays that information asymmetry has a significant positive relationship with sales growth for Ghana. Column (4) presents that firm age has a significant negative relationship with sales growth for Kenya. Column (5) presents that AI is likely to increase the sales growth in Lesotho. Whereas artificial intelligence may reduce the sales growth in the same country, Lesotho.

Table 10. Sales growth and AI by country.

	(1)	(2)	(3)	(4)	(5)
	Central African Republic	Ethiopia	Ghana	Kenya	Lesotho
Variables	Sales growth	Sales growth	Sales growth	Sales growth	Sales growth
AI index	−3.95e-05	0.0105***	−0.0721	0.00111	5.84e-05*
AI index	(6.54e-05)	(0.00272)	(0.0922)	(0.00149)	(3.49e-05)
Firm age	−1.44e-06	−0.000570***	0.00490	−0.000339***	3.62e-06
Firm age	(1.04e-05)	(0.000147)	(0.00966)	(9.37e-05)	(4.89e-06)
IA index	0.000467	−0.00528*	0.159*	−0.00183	−8.98e-05*
IA index	(0.00105)	(0.00282)	(0.0926)	(0.00224)	(5.03e-05)
Legal status	−5.10e-05	0.0198**	0.210	−0.00565	7.70e-05
Legal status	(0.000402)	(0.00846)	(0.338)	(0.00447)	(0.000114)
Constant	0.000530	−0.594***	−0.696***	−0.349***	0.000156*
Constant	(0.000481)	(0.00364)	(0.155)	(0.00266)	(8.80e-05)
Observations	78	902	504	1,902	125
R-squared	0.011	0.036	0.006	0.014	0.036

*** p < 0.01,

** p < 0.05,

* p < 0.1; IA index denotes information asymmetric index; AI implies artificial intelligence; Legal status means company with shareholders and without shareholders Source: computed by the Author 2026

Table 11 column (1) presents that AI has a significant positive relationship with sales growth in Nigeria. Having legal status (shareholder involvement) may likely increase sales growth in the Nigerian case. Column (2) presents that firm age has a significant negative relationship with sales growth for Rwanda. Column (3) displays that legal status has a significant negative relationship with sales growth for Sierra Leone. Column (4) presents that AI has a significant negative relationship with sales growth in South Africa. Similarly, firm age has a significant positive relationship with sales growth. Information asymmetry has a significant positive relationship with sales growth in South Africa. Column (5) presents that AI has an insignificant negative relationship with sales growth for Tanzania.

Table 11. Sales growth and AI by country.

	(1)	(2)	(3)	(4)	(5)
	Nigeria	Rwanda	Sierra Leone	South Africa	Tanzania
Variables	Sales growth	Sales growth	Sales growth	Sales growth	Sales growth
AI	0.218***	−0.000151	−0.00270	−0.00264**	9.42e-06
AI	(0.0437)	(0.000187)	(0.00267)	(0.00108)	(1.19e-05)
Firm age	6.16e-06	−2.26e-05**	0.000227	−0.000253***	9.80e-07
Firm age	(0.00170)	(1.07e-05)	(0.000277)	(4.24e-05)	(7.30e-07)
IA index	−0.0203	−8.39e-05	−0.000277	0.0129***	1.27e-05
IA index	(0.0260)	(0.000200)	(0.00311)	(0.00121)	(1.88e-05)
Legal status	0.553**	0.000255	−0.0486*	0.00160	−5.39e-05
Legal status	(0.261)	(0.000612)	(0.0278)	(0.00953)	(6.95e-05)
Constant	0.258***	−0.998***	−0.0839***	−0.0511***	0.000130***
Constant	(0.0279)	(0.000292)	(0.00479)	(0.00148)	(1.88e-05)
Observations	5,520	300	327	1,808	423
R-squared	0.018	0.008	0.052	0.090	0.008

*** p < 0.01,

** p < 0.05,

* p < 0.1; IA index denotes information asymmetric index; AI implies artificial intelligence; Legal status means company with shareholders and without shareholders Source: computed by the Author 2026

Furthermore, the study examines the industry to determine whether contributions from specific industries drive labour productivity growth. Table 12 column (1) shows that AI and information asymmetry have a significant positive relationship with labour productivity growth, but fabricated metal has an insignificant negative relationship with labour productivity growth. Column (2) revealed that information asymmetry has a significant positive relationship with labour productivity growth, but Food and beverages have an insignificant positive relationship with labour productivity growth. Column (3) displays that AI and information asymmetry have a significant positive relationship with labour productivity growth, and furniture has a significant positive relationship with labour productivity growth. Column (4) shows that AI and information asymmetry have a significant positive relationship with labour productivity growth, but garment has a significant negative relationship with labour productivity growth. Column (5) shows that AI and information asymmetry have a significant positive relationship with labour productivity growth, whereas non-metallic minerals have an insignificant negative relationship. Column (6) shows that AI and information asymmetry have a significant positive relationship with labour productivity growth, whereas publishing has an insignificant positive relationship.

Table 12. Labour Productivity Growth and AI by industry.

	(1)	(2)	(3)	(4)	(5)	(6)
Variables	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth
AI	0.102***	0.0559	0.139***	0.0626***	0.0782***	0.137***
AI	(0.0206)	(0.0389)	(0.0292)	(0.0142)	(0.0181)	(0.0291)
Firm age	−0.00175***	−0.00479***	0.000388	−0.00322***	−0.00461***	0.000281
Firm age	(0.000549)	(0.00171)	(0.00129)	(0.000425)	(0.000678)	(0.00128)
IA index	0.0350*	0.0968**	0.0294	0.0298*	0.0363*	0.0315
IA index	(0.0200)	(0.0420)	(0.0254)	(0.0163)	(0.0207)	(0.0262)
Fabricated metal	−0.0354
Fabricated metal	(0.0598)
Food & beverages		0.0369
Food & beverages		(0.138)
Furniture			0.552**
Furniture			(0.238)
Garment				−0.0932***
Garment				(0.0295)
Nonmetallic mineral					−0.0793
Nonmetallic mineral					(0.0651)
Publishing						0.183
Publishing						(0.186)
Constant	0.139***	0.181***	0.142***	0.0781***	0.110***	0.149***
Constant	(0.0154)	(0.0461)	(0.0226)	(0.0132)	(0.0164)	(0.0225)
Observations	8,208	11,010	6,401	10,109	8,302	6,401
R-squared	0.009	0.002	0.015	0.008	0.009	0.011

*** p < 0.01,

** p < 0.05,

* p < 0.1; Source: computed by the Author 2026

Table 13 column (1) shows that information asymmetry has a significant positive relationship with labour productivity growth, but other manufacturing has an insignificant positive relationship with labour productivity growth. Column (2) revealed that information asymmetry has a significant positive relationship with labour productivity growth, but other services have an insignificant negative relationship with labour productivity growth. Column (3) shows that AI and information asymmetry are positively and significantly associated with labour productivity growth. Also, motor vehicle repair sales have a significant positive relationship with labour productivity growth. Column (4) shows that AI and information asymmetry have a significant positive relationship with labour productivity growth, but wholesales have a significant negative relationship with labour productivity growth. Column (5) shows that information asymmetry has a significant positive relationship with labour productivity growth, whereas hotels and restaurants show an insignificant positive relationship.

Table 13. Labour Productivity Growth and AI by Industry.

	(1)	(2)	(3)	(4)	(5)
Variables	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth
AI	0.0418	0.0453	0.136***	0.137***	0.0581
AI	(0.0363)	(0.0353)	(0.0290)	(0.0292)	(0.0474)
Firm age	−0.00586***	−0.00504***	0.000349	0.000356	−0.00612**
Firm age	(0.00182)	(0.00154)	(0.00129)	(0.00129)	(0.00249)
IA index	0.0915**	0.0928**	0.0322	0.0326	0.121**
IA index	(0.0402)	(0.0403)	(0.0262)	(0.0262)	(0.0540)
Other manufacturing	0.287
Other manufacturing	(0.186)
Other services		−0.0674
Other services		(0.0477)
Repair sales of motor			0.725**
Repair sales of motor			(0.341)
Wholesales				−0.182*
Wholesales				(0.103)
Hotel & restaurant					−0.0122
Hotel & restaurant					(0.0722)
Constant	0.176***	0.202***	0.143***	0.154***	0.231***
Constant	(0.0404)	(0.0490)	(0.0219)	(0.0222)	(0.0646)
Observations	11,837	11,837	6,401	6,401	9,281
R-squared	0.003	0.002	0.016	0.011	0.002

*** p < 0.01,

** p < 0.05,

* p < 0.1; Source: computed by the Author 2026

Table 14 column (1) shows that information asymmetry has a significant positive relationship with labour productivity growth, but retail has a significant negative relationship with labour productivity growth (LPG). Column (2) revealed that information asymmetry has a significant positive relationship with labour productivity growth, while transportation has a significant negative relationship with labour productivity growth. Column (3) displays that AI and information asymmetry have a significant positive relationship with labour productivity growth, and the manufacturing panel has a significant positive relationship with labour productivity growth. Column (4) shows that AI and information asymmetry have a significant positive relationship with labour productivity growth, and the retail panel has a significant positive relationship with labour productivity growth. Column (5) shows that AI and information asymmetry have a significant positive relationship with labour productivity growth, while the other service panel also exhibits a significant positive relationship with labour productivity growth.

Table 14. Labour Productivity Growth and AI by industry.

	(1)	(2)	(3)	(4)	(5)
Variables	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth	Labour Productivity Growth
AI	0.0462	0.102***	0.0960***	0.130***	0.135***
AI	(0.0382)	(0.0206)	(0.0208)	(0.0289)	(0.0289)
Firm age	−0.00515***	−0.00164***	−0.00191***	0.000210	−9.66e-05
Firm age	(0.00159)	(0.000549)	(0.000550)	(0.00129)	(0.00130)
IA index	0.0942**	0.0333*	0.0366*	0.0345	0.0254
IA index	(0.0418)	(0.0197)	(0.0198)	(0.0261)	(0.0263)
Retails	−0.127***
Retails	(0.0479)
Transportation		−0.157***
Transportation		(0.0440)
Manufacturing panel			0.338***
Manufacturing panel			(0.105)
Retail panel				0.542***
Retail panel				(0.145)
Other service panel					0.706***
Other service panel					(0.204)
Constant	0.216***	0.140***	0.127***	0.139***	0.143***
Constant	(0.0497)	(0.0153)	(0.0150)	(0.0218)	(0.0218)
Observations	11,309	8,208	8,208	6,401	6,401
R-squared	0.002	0.010	0.014	0.017	0.019

*** p < 0.01,

** p < 0.05,

* p < 0.1; Source: computed by the Author 2026

4. Conclusion

The empirical analysis offers insight into how AI can affect labour productivity and sales growth in sub-Saharan African countries, utilising a WBES dataset. The study selected 10 countries in sub-Saharan Africa, which are growing economies in terms of gross domestic product (GDP). We generated an AI variable from the proxy of innovation-related variables in the enterprise survey dataset. The current study investigated the relationships between artificial intelligence (AI) and labour productivity growth, as well as between AI and firm-level sales growth. The results revealed that AI is an enabler of productivity improvement^1,14 through capacity building, professional advancement, and understanding of timely technological advancements. It is noteworthy that harmonising investment in capacity building and enabling innovation in organisations are prerequisites for the AI and labour productivity outcomes. When firms move in tandem with technological advancements and creativity in their production. Firms’ innovation in goods and services is likely to attract customers, leading to a spillover effect on sales growth.

However, the effects of AI on labour productivity and sales growth vary by firm size. The findings show that micro-scale enterprises make the highest contributions to labour productivity⁴⁴ and sales growth.¹⁴ The results show an upward-downward trend among firms, with large firms making the least contribution to labour productivity in Africa, followed by medium-scale enterprises. The contribution of small-scale enterprises was higher than that of medium and large-scale enterprises. Also, the results show that AI, labour productivity, and sales growth yield diverse outcomes across the selected countries. Counties such as Ethiopia and Nigeria have a significant positive relationship between their AI and labour productivity.^45,46 One limitation of this study is that some countries with a large number of observations might determine the outcomes. AI has a significant positive relationship with labour productivity and sales growth; however, the outcome varies across industries. Industries such as furniture, motor vehicle repair sales, manufacturing, retail, and other service sectors influence labour productivity growth.

Overall, the results underline the strategic importance of AI not only as a technological upgrade but as a catalyst for firm performance and growth in sub-Saharan African countries. With a large population in Africa, investment in AI is likely to be a lucrative opportunity in a rapidly changing environment. The study suggests government policy recommendations and industry strategies that facilitate a smooth transition of AI-driven initiatives and foster their success. The study could not identify a relevant instrument (IV) to control endogeneity. The study made a significant contribution to the literature by improving our understanding of AI and labour productivity using firm-level datasets and by employing techniques that control for unobserved heterogeneity, eliminate bias from omitted-variable bias within groups, and account for standard errors. Future research should investigate the longitudinal impacts, sectoral differences, and inclusive AI deployment on shaping sustainable firm outcomes.

This project contains the following underlying data:

Enterprise survey dataset (firm-level data). The study uses a panel dataset from 10 selected countries, including the Central African Republic, Ethiopia, Ghana, Kenya, Lesotho, Nigeria, Rwanda, Sierra Leone, South Africa, and Tanzania, covering 2007 to 2024. Contains information on firm characteristics, labour productivity, sales, and business environment indicators across Sub-Saharan African countries.

Processed dataset used for analysis. Variables were sorted for each country before setting a data panel. The data processing includes cleaned and transformed variables used to estimate labour productivity and sales growth.

Replication files. Contains the dataset used to compute summary statistics and generate tables and figures presented in the study.

The data are publicly available from the World Bank Enterprise Surveys portal and can be accessed upon registration. The processed dataset is available at https://doi.org/10.6084/m9.figshare.32055780,⁴⁷ and the replication material used in this study is available from the corresponding author.

The cleaned data is available under the terms of the Figshare.com metadata access policy (compatible with CC BY 4.0 use).

Data availability

Name of data repository: figshare.com

Title of dataset: Estimation of Firm Labour Productivity and Sales Growth from Artificial Intelligence in Sub-Saharan African Countries.

Persistent identifier: https://doi.org/10.6084/m9.figshare.32055780.⁴⁷

References

1. Damioli G, Van Roy V, Vertesy D: The impact of artificial intelligence on labor productivity. Eurasian Bus. Rev. 2021; 11: 1–25. Publisher Full Text
2. Czarnitzki D, Fernández GP, Rammer C: Artificial intelligence and firm-level productivity. J. Econ. Behav. Organ. 2023; 211: 188–205. Publisher Full Text
3. SenGupta S: Labour market outcomes of artificial intelligence: A firm-level comparative study of advanced and emerging market economies using local projections approach. Available at SSRN 5176353. 2025.
4. Sohn K, Kwon O: Technology acceptance theories and factors influencing artificial Intelligence-based intelligent products. Telematics Inform. 2020; 47: 101324. Publisher Full Text
5. Nguyen TL, Nguyen VP, Dang TVD: Critical factors affecting the adoption of artificial intelligence: An empirical study in Vietnam. The Journal of Asian Finance, Economics and Business. 2022; 7(5): 225–236. Publisher Full Text
6. Kelly S, Kaye SA, Oviedo-Trespalacios O: What factors contribute to the acceptance of artificial intelligence? A systematic review. Telematics Inform. 2023; 77: 101925. Publisher Full Text
7. Wang G, Obrenovic B, Gu X, et al.: Fear of the new technology: Investigating the factors that influence individual attitudes toward generative Artificial Intelligence (AI). Curr. Psychol. 2025; 44: 8050–8067. Publisher Full Text
8. Change WWV: Labor Displacement in Artificial Intelligence Era: A Systematic Literature Review. Taiwan Journal of East Asian Studies. 2020; 17(2).
9. Occhipinti JA, Hynes W, Prodan A, et al.: Generative AI may create a socioeconomic tipping point through labour displacement. Sci. Rep. 2025; 15(1): 26050. PubMed Abstract | Publisher Full Text | Free Full Text
10. Bullemore-Campbell J, Tautiva JAD, Cristobal-Fransi E: AI in sales: Environmental, behavioral, and technological drivers of adoption in an emerging market. Social Sciences & Humanities Open. 2025; 12: 102161. Publisher Full Text
11. Jazairy A, Shurrab H, Chedid F: Impact pathways: walking a tightrope—unveiling the paradoxes of adopting artificial intelligence (AI) in sales and operations planning. Int. J. Oper. Prod. Manag. 2025; 45(13): 1–27. Publisher Full Text
12. Uwagaba J, Omotosho TD, George GO: Exploring the barriers to artificial intelligence adoption in Sub-Saharan Africa’s Small and Medium Enterprises and the potential for increased productivity. World Wide Journal of Multidisciplinary Research and Development. 2023.
13. Maswana JC: Exploring the growth effects of Artificial Intelligence in developing countries in Africa using a semi-endogenous growth model. American Economic Association/ASSA Conference, San Antonio, Tx. 2024 January 6.
14. Gao X, Feng H: AI-driven productivity gains: Artificial intelligence and firm productivity. Sustainability. 2023; 15(11): 8934. Publisher Full Text
15. Kádárová J, Fiľo M, Sukopová D, et al.: Artificial Intelligence Adoption and Labour Productivity in Slovakia and the EU27: Implications for Sustainable Economic Growth. Sustainability. 2026; 18(4): 2135. Publisher Full Text
16. De la Peña N, Granados OM: Artificial intelligence solutions to reduce information asymmetry for Colombian cocoa small-scale farmers. Information Processing in Agriculture. 2024; 11(3): 310–324. Publisher Full Text
17. Bany Salameh H, Elrefae G, Eletter S: Artificial intelligence and information symmetry: a path to mitigate adverse selection. Clust. Comput. 2025; 28(9): 601. Publisher Full Text
18. Chatterjee S, Chaudhuri R, Vrontis D, et al.: Digital workplace and organization performance: Moderating role of digital leadership capability. J. Innov. Knowl. 2023; 8(1): 100334. Publisher Full Text
19. Bullemore-Campbell J, Tautiva JAD, Cristobal-Fransi E: AI in sales: Environmental, behavioral, and technological drivers of adoption in an emerging market. Social Sciences & Humanities Open. 2025; 12: 102161. Publisher Full Text
20. Solow RM: Growth theory and after. Am. Econ. Rev. 1988; 78(3): 307–317.
21. Tingbani I, Salia S, Hartwell CA, et al.: Looking in the rear-view mirror: Evidence from artificial intelligence investment, labour market conditions and firm growth. Int. J. Financ. Econ. 2025; 30(1): 961–982. Publisher Full Text
22. Lahiri S, Bose I, Majumdar A: Rebel Foods' Cloud Kitchen Technologies: Food for Thought?. COMMUNICATIONS OF THE ASSOCIATION FOR INFORMATION SYSTEMS; 2024.
23. Dinh TL, Vu MC, Tran GTC: Artificial Intelligence in SMEs: Enhancing Business Functions Through Technologies and Applications. INFORMATION; 2025.
24. Chen L, Jiang M, Jia F, et al.: Artificial intelligence adoption in business-to-business marketing: toward a conceptual framework. J. Bus. Ind. Mark. 2022; 37(5): 1025–1044. Publisher Full Text
25. Sivathanu B, Pillai R: Generative AI: fuelling e-commerce revenue growth. Journal of Revenue and Pricing Management. 2025; 1–16.
26. Chen J, Zhou W, Frankwick GL: Firm AI adoption intensity and marketing performance. J. Comput. Inf. Syst. 2025; 65(2): 172–189. Publisher Full Text
27. Bujor D, Ene AB: AI-Driven Predictive Customer Analytics for Forecasting Behavior, Churn and Future Buying Patterns. PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON BUSINESS EXCELLENCE. 2025; 19. : 981–994. Publisher Full Text
28. Ghazal TM, El Kadhi N, Ahmad M: Unlocking business insights with big data analytics and predictive AI: Discovering hidden patterns for accurate sales forecasting. International Journal of Advanced and Applied Sciences. 2025; 12: 237–245. Publisher Full Text
29. Tudose C, Boiangiu CA, Piele M: The impact of using AI and ChatGPT in organisations. University Politehnica of Bucharest, Scientific Bulletin Series C: Electrical Engineering and Computer Science; 2025.
30. Fan G: Influence of AI on business strategies of music culture communication companies. Transformations In Business & Economics. 2025; 24(2 (65)): 355–376. Publisher Full Text
31. Maheshwari P, Kamble S, Pundir A, et al.: Internet of things for perishable inventory management systems: an application and managerial insights for micro, small and medium enterprises. Ann. Oper. Res. 2025; 350(2): 395–423. PubMed Abstract | Publisher Full Text | Free Full Text
32. Neumeyer X, Santos SC, Morris MH: Overcoming barriers to technology adoption when fostering entrepreneurship among the poor: The role of technology and digital literacy. IEEE Trans. Eng. Manag. 2020; 68(6): 1605–1618. Publisher Full Text
33. Yang D, Zhao WG, Yang YM: Approaching Artificial Intelligence in business and economics research: a bibliometric panorama (1966-2020). Tech. Anal. Strat. Manag. 2024; 36: 563–578. Publisher Full Text
34. Mohammed I, Bunyaminu A: Major obstacles facing business enterprises in an emerging economy: the case of Ghana using the World Bank Enterprise Survey. J. Small Bus. Enterp. Dev. 2021; 28(3): 475–487. Publisher Full Text
35. Chukwuma-Ume N, Ume CO: Status of agribusiness enterprises in Nigeria: evidence from World Bank Enterprise Survey data. Journal of Agribusiness in Developing and Emerging Economies. 2025; 15(2): 445–467. Publisher Full Text
36. Khan A, Talukder MS, Islam QT, et al.: The impact of business analytics capabilities on innovation, information quality, agility and firm performance: The moderating role of industry dynamism. VINE J. Inf. Knowl. Manag. Syst. 2024; 54(5): 1124–1152. Publisher Full Text
37. Bhattacharya P, Rath BN: Innovation and firm-level labour productivity: A comparison of Chinese and Indian manufacturing based on enterprise surveys. Science, Technology and Society. 2020; 25(3): 465–481. Publisher Full Text
38. Xiao Z, Gao J, Wang Z, et al.: Power shortage and firm productivity: Evidence from the World Bank Enterprise Survey. Energy. 2022; 247: 123479. Publisher Full Text
39. Shekar KC: Informality, innovation, and firm performance: evidence from World Bank Enterprise Survey in India, 2013–2014. J. Innov. Entrep. 2025; 14(1): 17. Publisher Full Text
40. Esteve-Pérez S, Pieri F, Rodriguez D: Age and productivity as determinants of firm survival over the industry life cycle. Ind. Innov. 2018; 25(2): 167–198. Publisher Full Text
41. Cowling M, Tanewski G: On the productive efficiency of Australian businesses: firm size and age class effects. Small Bus. Econ. 2019; 53(3): 739–752. Publisher Full Text
42. Woltjer G, Van Galen M, Logatcheva K: Industrial innovation, labour productivity, sales and employment. Int. J. Econ. Bus. 2021; 28(1): 89–113. Publisher Full Text
43. Zhu X, Zuo X, Li H: The dual effects of heterogeneous environmental regulation on the technological innovation of Chinese steel enterprises—Based on a high-dimensional fixed effects model. Ecol. Econ. 2021; 188: 107113. Publisher Full Text
44. Mishra S, Ewing MT, Cooper HB: Artificial intelligence focus and firm performance. J. Acad. Mark. Sci. 2022; 50(6): 1176–1197. Publisher Full Text
45. Girmay FG: Artificial Intelligence for Ethiopia: Opportunities and Challenges. Information Technologist. 2019; 16(1).
46. Vaikosen B, Magini GU: Assessing the Impact of Artificial Intelligence on Productivity and Global Market Competitiveness: A Case Study of Nigeria. Faculty of Natural and Applied Sciences Journal of Computing and Applications. 2025; 2(2): 85–92.
47. Adediran O: Estimation of Firm Labour Productivity and Sales Growth from Artificial Intelligence in Sub-Saharan African Countries. [Data set]. Figshare. 2026. Publisher Full Text

Comments on this article Comments (0)

Version 1

VERSION 1 PUBLISHED 12 May 2026

Author details Author details

Sustainable Livelihoods, University of South Africa Graduate School of Business Leadership, Johannesburg, Gauteng, 1686, South Africa

Olanrewaju Adewole Adediran
Roles: Conceptualization, Data Curation, Formal Analysis, Investigation, Methodology, Software, Supervision, Validation, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing

Competing interests

No competing interests were disclosed.

Grant information

The author(s) declared that no grants were involved in supporting this work.

Article Versions (1)

version 1

Published: 12 May 2026, 15:711

https://doi.org/10.12688/f1000research.179912.1

Copyright

© 2026 Adewole Adediran O. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Download

Export To

metrics

	Views	Downloads
F1000Research	-	-
PubMed Central Data from PMC are received and updated monthly.	-	-

Citations

0

SEE MORE DETAILS

CITE

how to cite this article

Adewole Adediran O. Estimation of Firm Labour Productivity and Sales Growth from Artificial Intelligence in Sub-Saharan African Countries [version 1; peer review: 1 approved with reservations]. F1000Research 2026, 15:711 (https://doi.org/10.12688/f1000research.179912.1)

NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.

track

receive updates on this article

Track an article to receive email alerts on any updates to this article.

Open Peer Review

Version 1

VERSION 1

PUBLISHED 12 May 2026

Views

5

Reviewer Report 19 Jun 2026

Ruben Machucho Cadena, Polytechnic University of Victoria, Ciudad Victoria, Mexico

Approved with Reservations

https://doi.org/10.5256/f1000research.198472.r488903

The article examines how artificial intelligence (AI)–related innovation is associated with firm‑level labor‑productivity and sales growth in ten Sub‑Saharan African countries using panel data from the World Bank Enterprise Surveys (2007–2024). AI is proxied by an index constructed from innovation ... Continue reading

The article examines how artificial intelligence (AI)–related innovation is associated with firm‑level labor‑productivity and sales growth in ten Sub‑Saharan African countries using panel data from the World Bank Enterprise Surveys (2007–2024). AI is proxied by an index constructed from innovation and organizational questions, and information asymmetry by certification, listing, and audit indicators. Labor productivity (sales per worker) and sales growth are computed from current and three‑years‑prior values. The empirical strategy combines robust OLS, FGLS, and high‑dimensional fixed‑effects regressions with controls for age, size, foreign ownership, industry, and time. In pooled models, the AI index is positively associated with sales growth and, in some cases, with labor‑productivity growth, while firm age is generally negative and information asymmetry positive. Disaggregated results show stronger AI–performance associations for micro and small firms and substantial heterogeneity across countries and sectors.

Regarding the checklist:

- Presentation and literature – Partly. The paper is well structured and cites recent work, but the construct “AI index” needs clearer justification and consistent terminology, and the narrative around numerous country/industry regressions should be streamlined.

- Design and technical soundness – Partly. The panel‑data design and estimators are appropriate, but endogeneity between AI and performance, ambiguous formulas for growth variables, and lack of discussion of effect sizes and multiple testing must be addressed.

- Methods and detail – Partly. Data sources are clear and replication files are referenced, yet construction of the AI and information‑asymmetry indices, fixed‑effects structure, treatment of outliers, and sample‑selection rules require fuller documentation (e.g., in an appendix).

- Statistical analysis – Partly. Robust estimators are used, but results are interpreted too causally and differences across OLS/FGLS/HDFE are not systematically discussed.

- Data availability – Yes. WBES data and processed/replication materials are available through the World Bank and figshare.

- Conclusions – Partly. The core conclusion—positive associations between AI‑related capabilities and firm performance—is supported, but causal and speculative claims about AI as a “catalyst for growth” should be moderated.

Key must‑fix points: clarify and justify the AI measure; formalize growth‑variable definitions; explicitly confront endogeneity (e.g., with lagged AI and sensitivity checks) and reframe claims as correlational; and fully document econometric specifications and replication materials.

Is the work clearly and accurately presented and does it cite the current literature?

Partly
Is the study design appropriate and is the work technically sound?

Partly
Are sufficient details of methods and analysis provided to allow replication by others?

Partly
If applicable, is the statistical analysis and its interpretation appropriate?

Partly
Are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions drawn adequately supported by the results?

Partly

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Artificial intelligence and firm performance; Institutions, business environment, and information asymmetries in developing economies; Quantitative methods and panel-data econometrics in economic and management research

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

CITE

Report a concern

Respond or Comment

Comments on this article Comments (0)

Version 1

VERSION 1 PUBLISHED 12 May 2026

Open Peer Review

Reviewer Status

Reviewer Reports

	Invited Reviewers
	1
Version 1 12 May 26	read

Ruben Machucho Cadena, Polytechnic University of Victoria, Ciudad Victoria, Mexico

Comments on this article

All Comments(0)

Add a comment

Sign up for content alerts

Browse by related subjects

Back to all reports

Reviewer Report

5 Views

19 Jun 2026 | for Version 1

Ruben Machucho Cadena, Polytechnic University of Victoria, Ciudad Victoria, Mexico

5 Views Cite this report Responses(0)

Approved With Reservations

The article examines how artificial intelligence (AI)–related innovation is associated with firm‑level labor‑productivity and sales growth in ten Sub‑Saharan African countries using panel data from the World Bank Enterprise Surveys (2007–2024). AI is proxied by an index constructed from innovation and organizational questions, and information asymmetry by certification, listing, and audit indicators. Labor productivity (sales per worker) and sales growth are computed from current and three‑years‑prior values. The empirical strategy combines robust OLS, FGLS, and high‑dimensional fixed‑effects regressions with controls for age, size, foreign ownership, industry, and time. In pooled models, the AI index is positively associated with sales growth and, in some cases, with labor‑productivity growth, while firm age is generally negative and information asymmetry positive. Disaggregated results show stronger AI–performance associations for micro and small firms and substantial heterogeneity across countries and sectors.

Regarding the checklist:

- Presentation and literature – Partly. The paper is well structured and cites recent work, but the construct “AI index” needs clearer justification and consistent terminology, and the narrative around numerous country/industry regressions should be streamlined.

- Design and technical soundness – Partly. The panel‑data design and estimators are appropriate, but endogeneity between AI and performance, ambiguous formulas for growth variables, and lack of discussion of effect sizes and multiple testing must be addressed.

- Methods and detail – Partly. Data sources are clear and replication files are referenced, yet construction of the AI and information‑asymmetry indices, fixed‑effects structure, treatment of outliers, and sample‑selection rules require fuller documentation (e.g., in an appendix).

- Statistical analysis – Partly. Robust estimators are used, but results are interpreted too causally and differences across OLS/FGLS/HDFE are not systematically discussed.

- Data availability – Yes. WBES data and processed/replication materials are available through the World Bank and figshare.

- Conclusions – Partly. The core conclusion—positive associations between AI‑related capabilities and firm performance—is supported, but causal and speculative claims about AI as a “catalyst for growth” should be moderated.

Key must‑fix points: clarify and justify the AI measure; formalize growth‑variable definitions; explicitly confront endogeneity (e.g., with lagged AI and sensitivity checks) and reframe claims as correlational; and fully document econometric specifications and replication materials.

Is the work clearly and accurately presented and does it cite the current literature?

Partly
Is the study design appropriate and is the work technically sound?

Partly
Are sufficient details of methods and analysis provided to allow replication by others?

Partly
If applicable, is the statistical analysis and its interpretation appropriate?

Partly
Are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions drawn adequately supported by the results?

Partly

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Artificial intelligence and firm performance; Institutions, business environment, and information asymmetries in developing economies; Quantitative methods and panel-data econometrics in economic and management research

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Respond to this report

Responses (0)

[1] 1. Damioli G, Van Roy V, Vertesy D: The impact of artificial intelligence on labor productivity. Eurasian Bus. Rev. 2021; 11: 1–25. Publisher Full Text

[2] 2. Czarnitzki D, Fernández GP, Rammer C: Artificial intelligence and firm-level productivity. J. Econ. Behav. Organ. 2023; 211: 188–205. Publisher Full Text

[3] 3. SenGupta S: Labour market outcomes of artificial intelligence: A firm-level comparative study of advanced and emerging market economies using local projections approach. Available at SSRN 5176353. 2025.

[4] 4. Sohn K, Kwon O: Technology acceptance theories and factors influencing artificial Intelligence-based intelligent products. Telematics Inform. 2020; 47: 101324. Publisher Full Text

[5] 5. Nguyen TL, Nguyen VP, Dang TVD: Critical factors affecting the adoption of artificial intelligence: An empirical study in Vietnam. The Journal of Asian Finance, Economics and Business. 2022; 7(5): 225–236. Publisher Full Text

[6] 6. Kelly S, Kaye SA, Oviedo-Trespalacios O: What factors contribute to the acceptance of artificial intelligence? A systematic review. Telematics Inform. 2023; 77: 101925. Publisher Full Text

[7] 7. Wang G, Obrenovic B, Gu X, et al.: Fear of the new technology: Investigating the factors that influence individual attitudes toward generative Artificial Intelligence (AI). Curr. Psychol. 2025; 44: 8050–8067. Publisher Full Text

[8] 8. Change WWV: Labor Displacement in Artificial Intelligence Era: A Systematic Literature Review. Taiwan Journal of East Asian Studies. 2020; 17(2).

[9] 9. Occhipinti JA, Hynes W, Prodan A, et al.: Generative AI may create a socioeconomic tipping point through labour displacement. Sci. Rep. 2025; 15(1): 26050. PubMed Abstract | Publisher Full Text | Free Full Text

[10] 10. Bullemore-Campbell J, Tautiva JAD, Cristobal-Fransi E: AI in sales: Environmental, behavioral, and technological drivers of adoption in an emerging market. Social Sciences & Humanities Open. 2025; 12: 102161. Publisher Full Text

[11] 11. Jazairy A, Shurrab H, Chedid F: Impact pathways: walking a tightrope—unveiling the paradoxes of adopting artificial intelligence (AI) in sales and operations planning. Int. J. Oper. Prod. Manag. 2025; 45(13): 1–27. Publisher Full Text

[12] 12. Uwagaba J, Omotosho TD, George GO: Exploring the barriers to artificial intelligence adoption in Sub-Saharan Africa’s Small and Medium Enterprises and the potential for increased productivity. World Wide Journal of Multidisciplinary Research and Development. 2023.

[13] 13. Maswana JC: Exploring the growth effects of Artificial Intelligence in developing countries in Africa using a semi-endogenous growth model. American Economic Association/ASSA Conference, San Antonio, Tx. 2024 January 6.

[14] 14. Gao X, Feng H: AI-driven productivity gains: Artificial intelligence and firm productivity. Sustainability. 2023; 15(11): 8934. Publisher Full Text

[15] 15. Kádárová J, Fiľo M, Sukopová D, et al.: Artificial Intelligence Adoption and Labour Productivity in Slovakia and the EU27: Implications for Sustainable Economic Growth. Sustainability. 2026; 18(4): 2135. Publisher Full Text

[16] 16. De la Peña N, Granados OM: Artificial intelligence solutions to reduce information asymmetry for Colombian cocoa small-scale farmers. Information Processing in Agriculture. 2024; 11(3): 310–324. Publisher Full Text

[17] 17. Bany Salameh H, Elrefae G, Eletter S: Artificial intelligence and information symmetry: a path to mitigate adverse selection. Clust. Comput. 2025; 28(9): 601. Publisher Full Text

[18] 18. Chatterjee S, Chaudhuri R, Vrontis D, et al.: Digital workplace and organization performance: Moderating role of digital leadership capability. J. Innov. Knowl. 2023; 8(1): 100334. Publisher Full Text

[19] 19. Bullemore-Campbell J, Tautiva JAD, Cristobal-Fransi E: AI in sales: Environmental, behavioral, and technological drivers of adoption in an emerging market. Social Sciences & Humanities Open. 2025; 12: 102161. Publisher Full Text

[20] 20. Solow RM: Growth theory and after. Am. Econ. Rev. 1988; 78(3): 307–317.

[21] 21. Tingbani I, Salia S, Hartwell CA, et al.: Looking in the rear-view mirror: Evidence from artificial intelligence investment, labour market conditions and firm growth. Int. J. Financ. Econ. 2025; 30(1): 961–982. Publisher Full Text

[22] 22. Lahiri S, Bose I, Majumdar A: Rebel Foods' Cloud Kitchen Technologies: Food for Thought?. COMMUNICATIONS OF THE ASSOCIATION FOR INFORMATION SYSTEMS; 2024.

[23] 23. Dinh TL, Vu MC, Tran GTC: Artificial Intelligence in SMEs: Enhancing Business Functions Through Technologies and Applications. INFORMATION; 2025.

[24] 24. Chen L, Jiang M, Jia F, et al.: Artificial intelligence adoption in business-to-business marketing: toward a conceptual framework. J. Bus. Ind. Mark. 2022; 37(5): 1025–1044. Publisher Full Text

[25] 25. Sivathanu B, Pillai R: Generative AI: fuelling e-commerce revenue growth. Journal of Revenue and Pricing Management. 2025; 1–16.

[26] 26. Chen J, Zhou W, Frankwick GL: Firm AI adoption intensity and marketing performance. J. Comput. Inf. Syst. 2025; 65(2): 172–189. Publisher Full Text

[27] 27. Bujor D, Ene AB: AI-Driven Predictive Customer Analytics for Forecasting Behavior, Churn and Future Buying Patterns. PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON BUSINESS EXCELLENCE. 2025; 19. : 981–994. Publisher Full Text

[28] 28. Ghazal TM, El Kadhi N, Ahmad M: Unlocking business insights with big data analytics and predictive AI: Discovering hidden patterns for accurate sales forecasting. International Journal of Advanced and Applied Sciences. 2025; 12: 237–245. Publisher Full Text

[29] 29. Tudose C, Boiangiu CA, Piele M: The impact of using AI and ChatGPT in organisations. University Politehnica of Bucharest, Scientific Bulletin Series C: Electrical Engineering and Computer Science; 2025.

[30] 30. Fan G: Influence of AI on business strategies of music culture communication companies. Transformations In Business & Economics. 2025; 24(2 (65)): 355–376. Publisher Full Text

[31] 31. Maheshwari P, Kamble S, Pundir A, et al.: Internet of things for perishable inventory management systems: an application and managerial insights for micro, small and medium enterprises. Ann. Oper. Res. 2025; 350(2): 395–423. PubMed Abstract | Publisher Full Text | Free Full Text

[32] 32. Neumeyer X, Santos SC, Morris MH: Overcoming barriers to technology adoption when fostering entrepreneurship among the poor: The role of technology and digital literacy. IEEE Trans. Eng. Manag. 2020; 68(6): 1605–1618. Publisher Full Text

[33] 33. Yang D, Zhao WG, Yang YM: Approaching Artificial Intelligence in business and economics research: a bibliometric panorama (1966-2020). Tech. Anal. Strat. Manag. 2024; 36: 563–578. Publisher Full Text

[34] 34. Mohammed I, Bunyaminu A: Major obstacles facing business enterprises in an emerging economy: the case of Ghana using the World Bank Enterprise Survey. J. Small Bus. Enterp. Dev. 2021; 28(3): 475–487. Publisher Full Text

[35] 35. Chukwuma-Ume N, Ume CO: Status of agribusiness enterprises in Nigeria: evidence from World Bank Enterprise Survey data. Journal of Agribusiness in Developing and Emerging Economies. 2025; 15(2): 445–467. Publisher Full Text

[36] 36. Khan A, Talukder MS, Islam QT, et al.: The impact of business analytics capabilities on innovation, information quality, agility and firm performance: The moderating role of industry dynamism. VINE J. Inf. Knowl. Manag. Syst. 2024; 54(5): 1124–1152. Publisher Full Text

[37] 37. Bhattacharya P, Rath BN: Innovation and firm-level labour productivity: A comparison of Chinese and Indian manufacturing based on enterprise surveys. Science, Technology and Society. 2020; 25(3): 465–481. Publisher Full Text

[38] 38. Xiao Z, Gao J, Wang Z, et al.: Power shortage and firm productivity: Evidence from the World Bank Enterprise Survey. Energy. 2022; 247: 123479. Publisher Full Text

[39] 39. Shekar KC: Informality, innovation, and firm performance: evidence from World Bank Enterprise Survey in India, 2013–2014. J. Innov. Entrep. 2025; 14(1): 17. Publisher Full Text

[40] 40. Esteve-Pérez S, Pieri F, Rodriguez D: Age and productivity as determinants of firm survival over the industry life cycle. Ind. Innov. 2018; 25(2): 167–198. Publisher Full Text

[41] 41. Cowling M, Tanewski G: On the productive efficiency of Australian businesses: firm size and age class effects. Small Bus. Econ. 2019; 53(3): 739–752. Publisher Full Text

[42] 42. Woltjer G, Van Galen M, Logatcheva K: Industrial innovation, labour productivity, sales and employment. Int. J. Econ. Bus. 2021; 28(1): 89–113. Publisher Full Text

[43] 43. Zhu X, Zuo X, Li H: The dual effects of heterogeneous environmental regulation on the technological innovation of Chinese steel enterprises—Based on a high-dimensional fixed effects model. Ecol. Econ. 2021; 188: 107113. Publisher Full Text

[44] 44. Mishra S, Ewing MT, Cooper HB: Artificial intelligence focus and firm performance. J. Acad. Mark. Sci. 2022; 50(6): 1176–1197. Publisher Full Text

[45] 45. Girmay FG: Artificial Intelligence for Ethiopia: Opportunities and Challenges. Information Technologist. 2019; 16(1).

[46] 46. Vaikosen B, Magini GU: Assessing the Impact of Artificial Intelligence on Productivity and Global Market Competitiveness: A Case Study of Nigeria. Faculty of Natural and Applied Sciences Journal of Computing and Applications. 2025; 2(2): 85–92.

[47] 47. Adediran O: Estimation of Firm Labour Productivity and Sales Growth from Artificial Intelligence in Sub-Saharan African Countries. [Data set]. Figshare. 2026. Publisher Full Text

Estimation of Firm Labour Productivity and Sales Growth from Artificial Intelligence in Sub-Saharan African Countries

Abstract

Keywords

1. Introduction

2. Data and methods

Table 1. Descriptive Statistics.

2.1 Methods

3. Results and interpretation

Table 2. Matrix of correlations.

Table 3. Variance inflation factor.

Table 4. Labour productivity growth and AI: robust OLS.

Table 5. Sales growth and AI: robust OLS.

Table 6. Labour productivity and AI using the HDFE estimator.

Table 7. Sales growth and AI using the HDFE estimator.

Table 8. Labour productivity and AI by country.

Table 9. Labour productivity and AI by country.

Table 10. Sales growth and AI by country.

Table 11. Sales growth and AI by country.

Table 12. Labour Productivity Growth and AI by industry.

Table 13. Labour Productivity Growth and AI by Industry.

Table 14. Labour Productivity Growth and AI by industry.

4. Conclusion

This project contains the following underlying data:

Data availability

References

Comments on this article Comments (0)

Open Peer Review

Comments on this article Comments (0)

Open Peer Review

Reviewer Status

Reviewer Reports

Comments on this article

Browse by related subjects

Competing Interests Policy

Stay Updated