In 2018, South Korea reduced a long-standing tax credit for automation investments. This tax credit reduction increases tax costs, is a negative tax incentive on robot installations and resembles the introduction of an indirect robot tax. We analyze the impact of this reform on corporate investment behavior and employment of South Korean firms. We use company-level data on South Korean firms and apply a difference-in-differences setting, exploiting both variation with regard to treatment and variation in robot densities across industries. We find consistent negative effects of the reform on investment and employment and positive effects on investment quality. Our results show first empirical implications of a robot tax for companies.

We use matched employer-employee data to study how tax policies that lower the cost of capital impact workers during periods of sectoral decline. The policy we study, bonus depreciation, was implemented by the United States government in 2001, coinciding with significant employment declines in manufacturing. Using difference-in-differences event study methodologies, we estimate the effects of firm and market-level exposure to the policy on long-run worker earnings and employment. We then explore how sectoral transformation mediates the effects of bonus depreciation on worker outcomes by exploring heterogeneity across worker occupation, demographics, and exposure to salient drivers of manufacturing decline.

We investigate firms’ decisions to claim investment tax incentives following the 2014 tax reform in Japan and their impacts on economic outcomes. Our novel instrumental variable, the ratio of tax-eligible capital input costs to total costs at the industry level, explained the variation in firms’ tax claims. Financially constrained firms and those with tax loss carryforwards claimed the tax incentives less often than their counterparts. Tax claimants increased capital expenditures compared to pre-claim levels, with this effect being pronounced among financially constrained firms. However, tax claiming did not improve overall productivity. While we observe a rise in labor productivity, this result was due to a reduction in employment, not an increase in value added. Our findings suggest that when a relatively small proportion of firms—less than 20% in our analysis—claims tax incentives, focusing on actual claimants shows larger effects of tax incentives compared to the intent-to-treat approach.

In this paper we examine crowding-in and crowding-out of public and private sector investments in response to quasi-experimental variation in vertical transfers received by European Union regions. Leveraging a threshold in the allocation rule of EU Cohesion Policy funds, one of the largest regional policies globally, we conduct a Regression Discontinuity Design. Our findings indicate a crowding-out effect on public investments, ranging from 17 to 45 cents per Euro spent, primarily driven by a shift towards current expenditures. However, this effect is outweighed by substantial crowding-in of private sector investments, notably in construction, services, and the finance sector. Complementary Difference-in-Differences estimation provides some evidence for intertemporal substitution. Our results speak to the significance of complementarites between public and private investments in shaping fiscal multipliers, which holds important implication for the efficient design of the policy.