Impact of Automation and Artificial Intelligence on Internal Macroeconomic Goals

Question

Increased use of automation and artificial intelligence is likely to affect the jobs and incomes of the lower-skilled workers significantly. However, it will also bring various other benefits to the economy.

Evaluate the impact of automation and artificial intelligence on an economy’s internal macroeconomic goals. [15]

Answer

An economy’s internal macroeconomic goals are actual and potential growth, price stability, low unemployment, and — increasingly emphasised — inclusive growth. The thesis is that automation and AI have a largely positive effect on these goals overall, but one whose net sign depends on the economy’s level of development and the policies in place to manage the displacement of lower-skilled labour.

Positive: higher actual and potential growth, with price stability. Automation and AI raise the quality of goods and services, raising foreign demand (X) and domestic demand (C); by raising productivity they raise the expected profitability of investment, attracting investment (I) and FDI. Together these raise aggregate demand and so output — actual growth. They also raise the quality of capital and other resources, raising productive capacity — potential growth.

Because aggregate demand and aggregate supply rise in tandem, growth can be achieved with price stability — inflationary pressure is contained. The rise in aggregate demand and output also raises the derived demand for labour, lowering demand-deficient unemployment, with new jobs concentrated in the technology sector as the economy undergoes a structural shift. One qualification: if the automation and AI are imported, import expenditure (M) rises, offsetting part of the rise in aggregate demand and so muting the actual-growth gain.

Negative: structural unemployment and reduced inclusive growth. Against this, automation substitutes capital for labour, making lower-skilled workers redundant. Because displaced workers cannot readily fill new vacancies — a skills mismatch — structural (technological) unemployment rises. Distribution worsens too: displaced workers lose regular income; those who remain in low-skilled jobs see stagnant wages; while demand for workers skilled in AI and automation rises, widening the income gap and undermining inclusive growth, which requires that all socioeconomic groups share in the gains. A government can blunt this with long-run supply-side measures such as retraining and reskilling grants (Singapore’s SkillsFuture), and the fall in demand-deficient unemployment from higher aggregate demand partly offsets the rise in structural unemployment.

Judgement. The overall impact is largely positive, but the net sign turns on two factors. The first is the level of development. A developed economy near full capacity — Singapore, with a tight labour market, or Japan, with an ageing population — benefits more, since automation supplements a constrained workforce and raises capacity with little unemployment cost. A developing economy reliant on labour-intensive industry, with much excess capacity, risks severe structural unemployment, so for it costs may outweigh benefits. The second is policy. A market-oriented government that does not intervene leaves free-market forces to generate structural unemployment; a government that deploys long-run supply-side policy — retraining and reskilling — can reduce displacement and upgrade its workforce. The reasoned conclusion is therefore that automation and AI are, on balance, beneficial to internal macroeconomic goals in developed economies with active reskilling policy, but potentially harmful in less-developed, labour-intensive economies lacking such policy — so whether they benefit or cost an economy depends decisively on its development stage and its policy response.