Humankind & Resource Distribution—XVIII: Technological Progress, Employment & Distribution of Productivity Gains

The preceding Part XVII examined knowledge as a  resource whose circulation can enlarge the productive capabilities of society. Its application to production raises a question that is becoming increasingly urgent: when technological progress enables human beings to produce more in less time, who acquires the additional income and who enjoys the time released from work? The answer determines whether innovation becomes an instrument of emancipation or another source of economic insecurity.

In 1930, John Maynard Keynes contemplated a future in which technological advancement and capital accumulation might make a fifteen-hour working week possible. His essay, Economic Possibilities for Our Grandchildren, was not simply a forecast about machinery. It invited reflection on the purposes for which economic progress should be pursued.[1] Nearly a century later, the question remains relevant even though the anticipated working week has not become the general condition of humanity.

Productivity is a relationship between output and inputs. Labour productivity, measured as output per hour worked, can increase through better machinery, improved skills, more efficient organisation or advances in knowledge. It does not follow that labour alone produced the increment, nor that the owners of machinery alone deserve it. Modern production combines accumulated scientific knowledge, public infrastructure, private investment, entrepreneurial judgement and human effort. The distribution of the resulting surplus is a separate institutional question.

Consider an enterprise employing one hundred workers for forty hours each week. If an innovation raises output per hour by 25 percent, the same physical output could, under simplified conditions, be produced in thirty-two hours per worker. Alternatively, the enterprise could maintain working hours and increase output by 25 per cent. It might also reduce employment. Which outcome occurs depends upon demand, investment costs, competition, contractual arrangements and bargaining power. The arithmetic establishes a possibility, not a prediction about actual firms.

The additional value may appear as higher wages, increased profits, lower consumer prices, additional tax revenue or shorter working hours. These channels can coexist. A reduction in prices may improve the purchasing power of households even when nominal wages remain unchanged. Higher profits may finance further investment. But where competition is weak and workers possess little bargaining power, productivity improvements can increase returns to ownership without proportionately improving the lives of employees.

Economic history cautions against assuming that technological change necessarily destroys employment. Mechanisation has eliminated particular occupations while generating others. The relevant distinction is between tasks performed by machines and the new tasks that emerge as production changes.

Daron Acemoglu and Pascual Restrepo distinguish the displacement of labour by automation from the restoration of labour demand through the creation of new tasks.[2] Neither process guarantees that displaced workers will obtain the new jobs, particularly when the skills required or the locations of employment differ.

Their empirical research on industrial robots in American labour markets found adverse effects on employment and wages in more exposed localities.[3] Such findings should not be extrapolated mechanically to every country or technology. They demonstrate, however, why an increase in national output cannot by itself establish that the transition has benefited every community. A region losing its principal industry may experience years of declining incomes even while consumers elsewhere enjoy cheaper products.

The distributional evidence deserves equal attention. The International Labour Organization’s Global Wage Report 2024–25 found that between 1999 and 2024 labour productivity in high-income countries increased by approximately 29 percent, compared with 15 percent growth in real wages. Much of the divergence occurred in the earlier years of that period.[4] These aggregate figures do not identify automation as the sole cause. They establish that productivity growth and wage growth need not proceed together.

The United Nations’ 2025 assessment of progress towards the Sustainable Development Goals reported that the global labour income share fell from 52.9 percent in 2015 to 52.3 percent in 2024.[5] This measure includes estimated labour earnings of the self-employed and must be distinguished from the wage share alone. Changes in sectoral composition, relative prices and measurement methods complicate interpretation. Nevertheless, the allocation of national income between labour and capital warrants systematic scrutiny alongside growth in gross domestic product.

Artificial intelligence introduces a further complication. Unlike earlier machinery primarily associated with physical tasks, it can assist or substitute for some activities involving language, analysis and information processing. Its effects will depend upon how enterprises reorganise work. A professional using technology to improve the quality of a service is experiencing a different transformation from an employee whose tasks are eliminated without a viable alternative. The same technical capability can support either arrangement.

Pakistan must approach this transformation with particular care. A country requiring more productive employment cannot assume that labour-saving investment will automatically absorb new entrants into its workforce. Nor can it preserve inefficient methods indefinitely merely to protect existing jobs. In agriculture, manufacturing and services, technological adoption needs to be accompanied by opportunities for acquiring skills, developing enterprises and entering new markets. Access to affordable electricity, connectivity and finance will influence who can participate.

The institutional response should begin before displacement occurs. Training linked to actual employment opportunities, portable social protection, collective negotiation and transitional income support can reduce the costs borne by workers. Profit-sharing and employee ownership may broaden participation in the returns to capital, although excessive concentration of workers’ savings in their own employers creates additional risks. Competition policy can help ensure that efficiency gains reach consumers rather than remaining exclusively with dominant enterprises.

Shorter working hours also deserve serious consideration. Reductions negotiated in sectors experiencing sustained productivity improvements could allow employees to retain earnings while gaining time for family responsibilities, education, cultural pursuits and civic participation. Such arrangements cannot be imposed uniformly without examining staffing requirements, costs and the circumstances of small enterprises. A shorter formal working week is of little benefit to a person compelled to undertake a second job to meet basic needs. Time security must accompany income security.

For research purposes, the proposition advanced is testable. Enterprises and sectors adopting comparable technologies could be examined over time to determine whether gains in real value added per hour are associated with higher median hourly earnings, shorter involuntary or excessive working hours, improved employment stability and lower consumer prices. Comparisons should account for changes in demand, capital intensity, workforce composition and inflation. Unpaid care and informal employment must not disappear from the assessment simply because they are difficult to measure. Evidence of adverse outcomes would require revision of the institutional arrangements, not rejection of the evidence.

The objective of technological progress cannot be confined to producing an ever-increasing volume of commodities. Its deeper significance lies in expanding the range of choices available to human beings. A machine that saves an hour of labour creates a possibility for freedom; the economic system determines whether that hour becomes leisure, additional production or unemployment. Shared prosperity requires institutions capable of converting technical possibilities into actual improvements in human life.

The next part will examine the ecological limits of economic expansion and the responsibilities of present generations towards those yet to come.

Notes

1. John Maynard Keynes, “Economic Possibilities for Our Grandchildren” (1930), in Essays in Persuasion.
2. Daron Acemoglu and Pascual Restrepo, “Automation and New Tasks: How Technology Displaces and Reinstates Labor”, Journal of Economic Perspectives, 33(2), 2019, pp. 3–30.
3. Daron Acemoglu and Pascual Restrepo, “Robots and Jobs: Evidence from US Labor Markets”, Journal of Political Economy, 128(6), 2020, pp. 2188–2244.
4. International Labour Organization, Global Wage Report 2024–25: Is Wage Inequality Decreasing Globally?, executive summary, 2024.
5. United Nations Statistics Division, The Sustainable Development Goals Extended Report 2025, Goal 10, indicator 10.4.1.

[To be continued]

Dr. Ikramul Haq, Advocate Supreme Court, Adjunct Faculty at Lahore University of Management Sciences (LUMS), member Advisory Board and Visiting Senior Fellow of Pakistan Institute of Development Economics (PIDE), holds an LLD in tax laws. He was full-time journalist from 1979 to 1984 with Viewpoint and Dawn. He also served Civil Services of Pakistan from 1984 to 1996.