This paper examines the heterogeneity of employment changes in the U.S. manufacturing sector from 1972 to 1986, focusing on gross job creation, destruction, and reallocation. The study uses a rich dataset with approximately 860,000 annual observations on 160,000 manufacturing establishments, providing a comprehensive view of establishment-level employment dynamics. Key findings include: 1. **Gross Job Creation and Destruction**: The average rates of gross job creation and destruction were 9.2% and 11.3% per year, respectively. These rates reflect high levels of job creation and destruction within narrowly defined sectors. 2. **Job Reallocation Rate**: The job reallocation rate, which measures the rate at which employment positions are reallocated across plants, was found to be significant, indicating that job reallocation is a major driver of worker reallocation. 3. **Persistence and Concentration**: Job creation and destruction were highly persistent, with average one-year persistence rates of 68% and 81%, respectively. Additionally, job destruction was concentrated among establishments that experienced dramatic growth rates, suggesting that the bulk of job destruction cannot be explained by normal worker attrition. 4. **Explanations for Simultaneous Job Creation and Destruction**: The study explores various theories to explain the coexistence of job creation and destruction, including passive learning about initial conditions, technical innovation, and differences in initial conditions or uncertainties about future conditions. The analysis reveals that these factors contribute to job reallocation, but the exact mechanisms remain complex. 5. **Variation by Plant Characteristics**: The job reallocation rate varied significantly by plant age, size, geographic region, and ownership type. Young, small, and single-unit plants exhibited higher job reallocation rates compared to older, larger, and multi-unit plants. The paper concludes that the magnitude and cyclical behavior of job reallocation are influenced by a combination of factors, including plant-level heterogeneity and macroeconomic conditions. The findings highlight the importance of job reallocation in understanding worker mobility and the dynamics of the labor market.This paper examines the heterogeneity of employment changes in the U.S. manufacturing sector from 1972 to 1986, focusing on gross job creation, destruction, and reallocation. The study uses a rich dataset with approximately 860,000 annual observations on 160,000 manufacturing establishments, providing a comprehensive view of establishment-level employment dynamics. Key findings include: 1. **Gross Job Creation and Destruction**: The average rates of gross job creation and destruction were 9.2% and 11.3% per year, respectively. These rates reflect high levels of job creation and destruction within narrowly defined sectors. 2. **Job Reallocation Rate**: The job reallocation rate, which measures the rate at which employment positions are reallocated across plants, was found to be significant, indicating that job reallocation is a major driver of worker reallocation. 3. **Persistence and Concentration**: Job creation and destruction were highly persistent, with average one-year persistence rates of 68% and 81%, respectively. Additionally, job destruction was concentrated among establishments that experienced dramatic growth rates, suggesting that the bulk of job destruction cannot be explained by normal worker attrition. 4. **Explanations for Simultaneous Job Creation and Destruction**: The study explores various theories to explain the coexistence of job creation and destruction, including passive learning about initial conditions, technical innovation, and differences in initial conditions or uncertainties about future conditions. The analysis reveals that these factors contribute to job reallocation, but the exact mechanisms remain complex. 5. **Variation by Plant Characteristics**: The job reallocation rate varied significantly by plant age, size, geographic region, and ownership type. Young, small, and single-unit plants exhibited higher job reallocation rates compared to older, larger, and multi-unit plants. The paper concludes that the magnitude and cyclical behavior of job reallocation are influenced by a combination of factors, including plant-level heterogeneity and macroeconomic conditions. The findings highlight the importance of job reallocation in understanding worker mobility and the dynamics of the labor market.