| Abstract: | We evaluate the use of firm-level union coverage rates in empirical models of firm behavior and performance. We focus on the potential for measurement error in both firm- and industry-level data, and find that firm-level union coverage rates provide more precise estimates of union effects. Higher union coverage at a firm is associated with slower employment and sales growth, decreased productivity in nonmanufacturing firms, increased productivity in manufacturing firms, lower profitability, and less investment in durable assets, such as research and development. |
