Economic Stability

Parameter

Economic Stability

Direction▲Higher is better

Current TrendMixed (productivity gains vs displacement risks)

MeasurementEmployment rates, inequality indices, transition costs

Metrics

Economic & Labor

Risks

Economic Disruption

Models

Overview

Economic Stability measures the resilience of economic systems to AI-driven changes—encompassing labor market adaptability, income distribution patterns, capital-labor balance, and the smoothness of economic transitions as AI transforms industries. Higher economic stability is better—it enables societies to capture AI’s benefits while managing disruptions that could otherwise fuel political instability or authoritarian responses.

AI development pace, policy responses, and market adaptation mechanisms all determine whether economic stability strengthens or weakens. Unlike simple employment metrics, this parameter captures the broader capacity of economic systems to absorb technological shocks while maintaining living standards and social cohesion.

This parameter underpins:

Social cohesion: Stable employment and income prevent social unrest
Political stability: Economic disruption fuels populism and instability
Investment capacity: Economic stability enables long-term planning and investment
Transition success: Smooth transitions allow workers to adapt without crisis

This framing enables:

Symmetric analysis: Tracking both destabilizing factors (rapid displacement) and stabilizing factors (new job creation)
Early warning: Detecting economic stress before it becomes crisis
Policy design: Crafting interventions that maintain stability during transition
Progress monitoring: Measuring adaptation capacity over time

Parameter Network

Loading diagram...

Contributes to: Societal Adaptability

Primary outcomes affected:

Transition Smoothness ↓↓↓ — Economic stability is the primary factor in smooth transitions

Current State Assessment

Global Exposure to AI Automation

Region	Jobs Exposed	High-Risk Share	Complementary Jobs	Key Sectors	Source
Advanced Economies	60%	25-30%	~30% (enhanced productivity)	Finance, admin, customer service	IMF 2024↗
United States	57% of work hours	40% highest exposure	Variable by sector	Content, data entry, translation	McKinsey 2025↗
European Union	55-65%	20-25%	25-35%	Manufacturing, services	WEF 2025↗
Emerging Markets	40%	15-20%	15-20%	Manufacturing, BPO	IMF 2024↗
Low-Income Countries	26%	8-12%	10-15%	Agriculture, basic services	IMF 2024↗
Global Average	40%	18-22%	~20%	Cross-sector	IMF 2024↗

Note: “Exposed” means AI can automate significant portions of job tasks; “High-Risk” means jobs where majority of tasks can be automated; “Complementary” means jobs where AI integration enhances rather than replaces workers. IMF research indicates roughly half of exposed jobs may benefit from AI integration, enhancing productivity, while the other half face potential displacement.

Labor Market Indicators (2024-2025)

Indicator	Current Value	Pre-AI Baseline (2019)	Trend
US Tech Employment Share	Declining since Nov 2022	Stable/growing	Worsening
Young Tech Worker Unemployment	+3 percentage points	Baseline	Rising
Freelance Writing Gigs	-42% since 2021	Baseline	Sharp decline
AI Job Creation	~120K direct jobs (2024)	~0	Growing
Net Job Impact (2024)	+107K net	N/A	Positive (early stage)

Sources: Goldman Sachs labor analysis↗, ITIF research, Challenger reports

Economic Inequality Trends

Metric	2019	2024	Projected 2030	Assessment	Source
Top 1% Income Share (US)	18.8%	19.5%	22-25%	Worsening	Goldman Sachs↗
Labor Share of GDP	58%	56%	50-54%	Declining	IMF 2024↗
Gini Coefficient (OECD avg)	0.32	0.33	0.35-0.38	Increasing	[505c3bc13c08e66a]
Within-Occupation Inequality	Baseline	Declining (2014-18)	Continued decline possible	Mixed signal	[505c3bc13c08e66a]
Median Wage Growth (real)	1.2%	0.8%	0.5-1.5%	Stagnating	Goldman Sachs↗

Note: OECD research (2014-2018) found AI reduced wage inequality within most occupations—consistent with findings that AI reduces productivity differentials between workers, with low performers benefiting most from AI tools. However, overall inequality continues rising due to other factors.

What “Healthy Economic Stability” Looks Like

Healthy economic stability during AI transition involves:

Gradual displacement: Automation pace matches retraining capacity
Broad-based gains: Productivity benefits shared across income levels
New job creation: Emerging roles absorb displaced workers
Adaptive institutions: Education, welfare, and labor systems evolve
Geographic distribution: Benefits not concentrated in AI hubs

Stability vs. Disruption Indicators

Healthy Stability	Dangerous Disruption
Unemployment rise < 2% annually	Unemployment surge > 5% annually
Retraining programs functional	Retraining overwhelmed
New job categories emerging	Jobs disappearing faster than emerging
Inequality growth < 0.01 Gini/year	Inequality growth > 0.03 Gini/year
Wage growth positive	Real wage decline
Social mobility maintained	Social mobility declining

Factors That Decrease Economic Stability (Threats)

Loading diagram...

Displacement Speed Factors

Threat	Mechanism	Evidence	Severity
Capability acceleration	AI advances faster than adaptation	McKinsey: 57% automatable now↗	High
Multi-sector simultaneity	Many industries disrupted at once	Customer service, content, admin hit together	High
Retraining limits	Workers can’t adapt fast enough	Brookings: retraining often fails↗	High
Geographic concentration	AI hubs benefit, other areas decline	Tech job concentration in few metros	Medium

Capital-Labor Shift

When AI substitutes for human labor across many domains, economic value increasingly flows to capital (AI owners) rather than labor (workers):

Dynamic	Current State	Trajectory	Risk
Labor share of GDP	56% (down from 65% in 1970)	Declining	High
Firm concentration	Top 4 tech firms: $10T+ market cap	Accelerating	High
Wage-productivity gap	Widening since 1979	Accelerating	High
Automation returns	Accruing primarily to capital owners	Accelerating	High

Winner-Take-All Dynamics

Factor	Mechanism	Current Example
Network effects	First-movers capture market	OpenAI/Anthropic/Google dominance
Data advantages	More users = better AI = more users	ChatGPT’s 100M+ users
Talent concentration	Top labs attract best researchers	<20 orgs can train frontier models
Compute barriers	$100M+ training runs exclude most	Only well-funded labs can compete

Factors That Increase Economic Stability (Supports)

New Job Creation

Category	Estimated Jobs	Timeline	Evidence	Confidence
AI development/maintenance	2-5M globally	2025-2030	Direct industry growth	High
AI training/prompt engineering	1-3M	2024-2027	Emerging occupation data	Medium
Human-AI collaboration roles	10-20M	2025-2035	WEF 2025: net +78M jobs by 2030↗	Medium-Low
Care economy expansion	15-30M	2025-2040	Aging populations, AI-resistant	Medium
Creative/artisanal premium	5-10M	2025-2035	”Made by humans” value	Low
Agriculture/delivery workers	8-15M	2025-2030	[ad99f84cc63f17f9]	High

Note: WEF Future of Jobs Report 2025 projects 170M new roles created globally by 2030, with 92M displaced—net gain of 78M jobs. However, this represents aggregate numbers; geographic and skill mismatches mean displaced workers may not fill new roles.

Policy Interventions

Intervention	Mechanism	Status	Effectiveness	Cost Estimate
Universal Basic Income	Decouples income from employment	160+ pilots globally since 1980s	Mixed (reduces poverty, health gains; employment effects unclear)	[ef2248e0ed39ef39]
AI Automation Tax	Tax companies replacing workers with AI	Proposed by Gates (2017), renewed interest 2024-25	Untested	Potentially $200-500B annually
Negative Income Tax	Targeted support for low earners	Proposed in various forms	Theoretical	$300-600B annually (US)
Transition assistance	Short-term support during retraining	Germany’s Kurzarbeit model	Moderate success	$50-100B annually
Education reform	Prepare workers for AI economy	Singapore’s SkillsFuture; [d27e126d8a8d6efb]	Early implementation	$100-200B annually (global)
Portable benefits	Benefits not tied to single employer	Gig economy proposals	Limited adoption	$20-50B annually

Note: UBI feasibility depends on AI productivity gains. Research suggests AI capability threshold for economically viable UBI could be reached between 2028 (rapid progress) and mid-century (slow progress). Current US GDP ($29T) and federal revenue ($4.9T) insufficient without significant tax reform.

Market Adaptation Mechanisms

Mechanism	How It Stabilizes	Current State
Wage adjustment	Lower wages attract hiring	Functioning but slow
Geographic mobility	Workers move to opportunity	Declining (housing costs)
Entrepreneurship	Displaced workers start businesses	30% of new businesses AI-related
Sector shift	Workers move to growing industries	Possible but friction-heavy

Gradual Capability Scaling

If AI capabilities advance gradually rather than rapidly, adaptation mechanisms have time to function:

Scenario	Displacement Rate	Adaptation Probability	Stability Impact
Slow capability scaling	2-3% workers/year	70-80%	Maintains stability
Moderate scaling	5-7% workers/year	40-60%	Strains stability
Rapid scaling	10%+ workers/year	20-30%	Threatens stability

Why This Parameter Matters

Consequences of Low Economic Stability

Domain	Impact	Severity	Historical Parallel
Social cohesion	Unrest, protests, crime increases	Critical	Great Depression, Rust Belt decline
Political stability	Populism, extremism, democratic erosion	Critical	1930s Europe, 2016 populist wave
Mental health	Depression, suicide, substance abuse	High	Deindustrialization regions
Investment climate	Uncertainty reduces long-term investment	High	Emerging market volatility
Human capital	Skill atrophy during prolonged unemployment	High	Long-term unemployment effects