Right-to-work laws boost state-level venture capital but appear to depress firms' AI investment, a pattern linked in part to lower wages; meanwhile, larger, cash-rich, highly exposed and high-ESG firms were most likely to exit Russia after the shock.

Empirical Essays on Labor Regulation, Geopolitical Shocks, and Investment

Sarkodie, Helena · January 13, 2026

openalex quasi_experimental medium evidence 7/10 relevance Summary only summary available; pdf_status=paywall DOI Source PDF

Structured author observations

Linked only from stored provider relations; the raw author line above is never matched by name.

OpenAlex

Latest observation: August 24, 2026

Helena Sarkodie provider ID

Semantic Scholar

Latest observation: August 24, 2026

Helena Sarkodie provider ID

Using difference-in-differences on RTW law timing and a firm-level divestment dataset, the thesis shows RTW laws raise state VC investment but reduce firms' AI investment (partly via lower wages), while larger, more Russia-exposed, cash-rich, leveraged and higher-ESG firms were more likely to divest from Russia.

Citation observations

Cumulative provider counts captured on specific dates; providers are never combined.

0 cumulative citations

OpenAlex · Observed August 24, 2026

View corpus context

0 cumulative citations

Semantic Scholar · Observed August 24, 2026

View corpus context

This thesis consists of three self-contained yet interrelated empirical essays in corporate finance. The first and third essays examine the impact of Right-to-Work (RTW) laws on investment, focusing on venture capital (VC) and artificial intelligence (AI), respectively, while the second analyzes the determinants of corporate divestment from Russia. Together, these essays explore how institutional and geopolitical shocks shape corporate investment decisions, offering a unified perspective on the interaction between external environments and firm behavior. In the first essay, we contribute to the growing literature on RTW laws and their influence on state-level VC investment. Employing a difference-in-differences strategy, we find that the passage of RTW laws increases VC investment. These results are robust to concerns about omitted variable bias, reverse causality, and unobserved local economic conditions. The positive effect of RTW laws on VC investment is particularly pronounced in highly unionized and technologically advanced states. The second essay examines the determinants of corporate divestment from Russia using a unique dataset of divestment decisions of firms operating in Russia. We find that larger firms with higher sales in Russia as well as higher cash reserves and leverage, are more likely to divest from Russia. Additionally, firms with high Environmental, Social, and Governance (ESG) scores and substantial advertising expenditures (for U.S. firms only) exhibit a greater propensity to exit the Russian market. Our causal evidence is economically important, with elasticities for Firm Size, Russian sales, Cash, Leverage, and Social ESG score at 2.413, 0.468, 0.352, 0.477, and 0.962 which indicates that a 1% increase in these variables increases the probability of exiting Russia by 2.431%, 0.468%, 0.352%, 0.477% and 0.962%, respectively. The third essay uses a difference-in-differences approach to explore how right-to-work (RTW) laws influence firm investment in artificial intelligence (AI). We find that firms headquartered in RTW laws states invest less in AI. Investigating employee wages as a potential mechanism, we find that RTW laws lead to lower wages. Our results remain robust across multiple specifications, including an alternative AI investment measure, a stacked difference-in-differences framework, falsification tests, dynamic treatment effects, and a control group limited to neighboring states. The negative effect of RTW laws on AI investment persists when focusing exclusively on IT firms and when analyzing AI investment growth.

Summary

Main Finding

Institutional (right-to-work laws) and geopolitical (Russia invasion) shocks materially affect corporate investment decisions.
Passage of RTW laws increases state-level venture capital (VC) investment but reduces firm-level investment in artificial intelligence (AI).
Corporate divestment from Russia is driven by firm size, exposure to Russian sales, liquidity and leverage, and by reputational factors (high ESG scores and, for U.S. firms, advertising intensity).

Key Points

Essay 1 (RTW → VC):
- Using difference-in-differences, RTW law passage is associated with higher VC investment at the state level.
- Effects are strongest in states with high unionization and advanced technology sectors.
- Results are robust to concerns about omitted variables, reverse causality, and unobserved local conditions.
Essay 2 (Determinants of Russian divestment):
- Firms more likely to divest from Russia if they are larger, have greater sales exposure to Russia, hold more cash, or carry higher leverage.
- Reputational considerations matter: higher Social ESG scores (and higher advertising for U.S. firms) are associated with greater propensity to exit.
- Reported elasticities (probability of exit per 1% increase): Firm Size 2.413, Russian Sales 0.468, Cash 0.352, Leverage 0.477, Social ESG 0.962.
Essay 3 (RTW → AI investment):
- Firms headquartered in RTW states invest less in AI (difference-in-differences).
- Evidence points to lower employee wages under RTW as a mechanism reducing firm incentives to invest in AI.
- Findings are robust: alternative AI measures, stacked DID, falsification tests, dynamic effects, neighbor-state control groups, and when limiting to IT firms and to AI investment growth.

Data & Methods

Primary identification strategy for Essays 1 and 3: difference-in-differences (DID) exploiting timing of RTW law adoption across states.
- Robustness checks: placebo/falsification tests, dynamic treatment effects, alternative control groups (including neighboring states), stacked DID frameworks, and multiple AI investment measures.
- Mechanism testing for Essay 3: analysis of wage outcomes to link RTW to AI investment declines.
Essay 2 uses a unique, firm-level dataset of divestment decisions from Russia across firms operating there.
- Empirical approach: regression analysis linking divestment decisions to firm characteristics (size, Russian sales exposure, cash, leverage), ESG scores, and advertising (for U.S. firms).
- Elasticities reported to quantify economic magnitudes.
Across essays: attention to omitted variable bias, reverse causality, and local economic confounders through robustness checks and specification variants.

Implications for AI Economics

Institutions shape AI capital allocation heterogeneously:
- RTW laws may attract VC (potentially favoring startup formation and regional financing flows) while simultaneously reducing incumbent firms’ AI investment — implying divergent effects on AI development vs. deployment.
Labor-market effects matter for AI adoption:
- Lower wages under RTW reduce the economic incentive for firms to automate or invest in AI, altering the channel through which labor-market institutions affect technological adoption and diffusion.
Regional AI ecosystems:
- Policy-induced shifts in VC and firm-level AI investment can change where AI innovation and commercialization concentrate, affecting regional comparative advantage and regional inequality in AI-capable jobs.
Corporate responses to geopolitical shocks:
- The Russia divestment results indicate that both financial exposure and reputational concerns drive corporate exits — an important consideration for modeling geopolitical risk in firm investment and AI supply chains.
For researchers and policymakers:
- When evaluating AI policy or forecasting AI diffusion, incorporate institution-driven labor-cost changes and the distinction between VC flows (new ventures) and incumbent firm R&D/adoption.
- Future research should further unpack channels (e.g., how wage changes map to automation incentives across industries) and explore long-run outcomes of RTW-driven shifts in the composition of AI investment.

Assessment

Paper Typequasi_experimental Evidence Strengthmedium — The diff-in-diff design for the RTW analyses is a credible quasi-experimental approach and the author reports multiple robustness checks (stacked DiD, falsification, dynamics, alternative measures and sub-samples), which strengthens causal claims; however, remaining concerns include potential violations of parallel trends, unobserved concurrent state-level shocks, measurement of AI investment, and the divestment analysis is largely correlational (subject to omitted variable bias and selection into the divestment sample). Methods Rigormedium — The thesis applies standard and advanced empirical tools (DiD, stacked DiD, falsification tests, dynamic treatment effects, neighbor controls) and reports elasticities and heterogeneity analyses, indicating careful empirical work; but the causal leverage varies across essays (stronger for the RTW DiD pieces, weaker for the divestment regressions), and important details (pre-trend diagnostics, robustness to alternative control groups beyond neighboring states, measurement construction of AI investment) limit a high-rigor rating without seeing full robustness tables. SampleEssay 1: state-level panel data on venture capital investment across U.S. states over years that include changes in RTW law status. Essay 2: a unique firm-level dataset of corporate divestment decisions by firms operating in Russia, merged with firm financials (size, sales in Russia, cash, leverage), ESG scores, and advertising data for U.S. firms. Essay 3: firm-level panel of AI investment measures (primary and an alternative measure), matched to firms' headquarters state and RTW law timing; includes sub-samples for IT firms and growth analyses. Themesadoption labor_markets IdentificationEssays 1 and 3 use difference-in-differences exploiting cross-state, over-time variation in the timing of Right-to-Work (RTW) law passage to identify effects on state-level VC investment and firm-level AI investment respectively, with robustness checks (stacked DiD, dynamic effects, falsification tests, neighboring-state controls, alternative AI measures). Essay 2 analyzes firm-level divestment decisions from Russia using cross-sectional/time-varying regressions of divestment probability on firm financials, sales exposure to Russia, ESG scores, advertising (U.S. only) and other controls; identification relies on conditional correlations and controls rather than an experimental/quasi-experimental instrument. GeneralizabilityFindings on RTW effects are specific to the U.S. legal and institutional context and may not generalize to countries without similar labor law structures., State-level VC results may not extrapolate to other forms of financing or to later periods with different macro or technology cycles., The AI investment measure may be an imperfect proxy for actual AI adoption/productivity effects, limiting inference about downstream economic impacts., Divestment sample may over-represent public or large firms and may not reflect private firms or small multinationals, risking selection bias., Industry heterogeneity exists (effects differ for IT firms), so results may not hold uniformly across sectors., Time-period-specific shocks (e.g., the 2022 Russia invasion and broader geopolitical context) may limit applicability to other geopolitical episodes.

Claims (11)

Claim	Direction	Outcome	Confidence & Evidence	Details
The passage of Right-to-Work (RTW) laws increases venture capital (VC) investment at the state level. Innovation Output	positive	Venture capital (VC) investment (state-level)	Reading fidelity high Study strength medium	not reported 0.48
The positive effect of RTW laws on VC investment is particularly pronounced in highly unionized and technologically advanced states. Innovation Output	positive	Venture capital (VC) investment (state-level), heterogeneous by state characteristics	Reading fidelity high Study strength medium	not reported 0.48
Larger firms are more likely to divest (exit) from Russia. Market Structure	positive	Probability of exiting (divesting from) Russia	Reading fidelity high Study strength medium	1% increase in Firm Size increases the probability of exiting Russia by 2.431% 0.48
Firms with higher sales in Russia are more likely to divest (exit) from Russia. Market Structure	positive	Probability of exiting (divesting from) Russia	Reading fidelity high Study strength medium	1% increase in Russian sales increases the probability of exiting Russia by 0.468% 0.48
Firms with higher cash reserves are more likely to divest (exit) from Russia. Market Structure	positive	Probability of exiting (divesting from) Russia	Reading fidelity high Study strength medium	1% increase in Cash increases the probability of exiting Russia by 0.352% 0.48
Firms with higher leverage are more likely to divest (exit) from Russia. Market Structure	positive	Probability of exiting (divesting from) Russia	Reading fidelity high Study strength medium	1% increase in Leverage increases the probability of exiting Russia by 0.477% 0.48
Firms with higher Social ESG scores are more likely to divest (exit) from Russia. Market Structure	positive	Probability of exiting (divesting from) Russia	Reading fidelity high Study strength medium	1% increase in Social ESG score increases the probability of exiting Russia by 0.962% 0.48
For U.S. firms, substantial advertising expenditures are associated with a greater propensity to exit the Russian market. Market Structure	positive	Probability of exiting (divesting from) Russia (U.S. firms subsample)	Reading fidelity high Study strength medium	not reported 0.48
Firms headquartered in RTW-law states invest less in artificial intelligence (AI). Innovation Output	negative	Firm-level AI investment (amount or investment indicator)	Reading fidelity high Study strength medium	not reported 0.48
RTW laws lead to lower employee wages (consistent with a mechanism by which RTW reduces AI investment). Wages	negative	Employee wages	Reading fidelity high Study strength medium	not reported 0.48
The negative effect of RTW laws on AI investment persists when restricting the sample to IT firms and when analyzing AI investment growth. Innovation Output	negative	AI investment (levels and growth), restricted to IT firms	Reading fidelity high Study strength medium	not reported 0.48