Managers and Productivity Differences Nezih Guner, Andrii Parkhomenko and Gustavo Ventura RIDGE, 2014
Motivation
Motivation • Understanding cross-country income differences.
Motivation • Understanding cross-country income differences. Bulk of work indicates that productivity differences are key.
Motivation • Understanding cross-country income differences. Bulk of work indicates that productivity differences are key. • Growing body of work emphasizes differences in management practices as a sources of productivity differences across countries. • Bloom and Van Reenen (2011), Bloom, Sadun and Van Reenen (2013).
Motivation • Understanding cross-country income differences. Bulk of work indicates that productivity differences are key. • Growing body of work emphasizes differences in management practices as a sources of productivity differences across countries. • Bloom and Van Reenen (2011), Bloom, Sadun and Van Reenen (2013). • How to interpret differences in management practices?
What we do We use large data sets from high-income countries. We document:
What we do We use large data sets from high-income countries. We document: 1 Age-earnings profiles of managers are steeper than those for non-managers in most countries.
What we do We use large data sets from high-income countries. We document: 1 Age-earnings profiles of managers are steeper than those for non-managers in most countries. 2 Earnings of managers grow faster with age than earnings of non managers in countries with higher GDP per worker.
What we do • We develop a life-cycle, span-of-control model to interpret these facts:
What we do • We develop a life-cycle, span-of-control model to interpret these facts: • managers invest in their managerial abilities.
What we do • We develop a life-cycle, span-of-control model to interpret these facts: • managers invest in their managerial abilities. • Goods are used to produce new skills. Goods and current skills are complementary in skill production.
What we do • We develop a life-cycle, span-of-control model to interpret these facts: • managers invest in their managerial abilities. • Goods are used to produce new skills. Goods and current skills are complementary in skill production. • Steepness of age-earnings profiles depends on the incentives of managers to invest.
What we do • We develop a life-cycle, span-of-control model to interpret these facts: • managers invest in their managerial abilities. • Goods are used to produce new skills. Goods and current skills are complementary in skill production. • Steepness of age-earnings profiles depends on the incentives of managers to invest. • Distribution of managerial abilities is endogenous and affected by distortions.
What we do • We develop a life-cycle, span-of-control model to interpret these facts: • managers invest in their managerial abilities. • Goods are used to produce new skills. Goods and current skills are complementary in skill production. • Steepness of age-earnings profiles depends on the incentives of managers to invest. • Distribution of managerial abilities is endogenous and affected by distortions. • Amplification. • Differences in management quality reflect the fact that some individuals have (endogenously) better managerial skills.
What we do Calibrate this model to U.S. economy.
What we do Calibrate this model to U.S. economy. • What is effect of cross country differences in exogenous aggregate productivity levels? Effects on relative managerial incomes, plant size and productivity measures.
What we do Calibrate this model to U.S. economy. • What is effect of cross country differences in exogenous aggregate productivity levels? Effects on relative managerial incomes, plant size and productivity measures. • Connect with literature on misallocation. What is role role of distortions (implicit output taxes)? • Restuccia and Rogerson (2008) and Guner, Ventura and Yi (2008) and others.
What we do Calibrate this model to U.S. economy. • What is effect of cross country differences in exogenous aggregate productivity levels? Effects on relative managerial incomes, plant size and productivity measures. • Connect with literature on misallocation. What is role role of distortions (implicit output taxes)? • Restuccia and Rogerson (2008) and Guner, Ventura and Yi (2008) and others. • Reallocation effect: resources (capital and labor) will move from more productive to less productive establishments.
What we do Calibrate this model to U.S. economy. • What is effect of cross country differences in exogenous aggregate productivity levels? Effects on relative managerial incomes, plant size and productivity measures. • Connect with literature on misallocation. What is role role of distortions (implicit output taxes)? • Restuccia and Rogerson (2008) and Guner, Ventura and Yi (2008) and others. • Reallocation effect: resources (capital and labor) will move from more productive to less productive establishments. • Skill investment effect: managers choose to invest less to improve their skills.
Managers and Non-Managers: Data • Use large cross sectional data sets for a group of high income countries. • 20 countries in our sample Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Iceland, Ireland, Israel, Italy, Luxembourg, Netherlands, Norway, Spain, Sweden, Switzerland, UK, US.
Managers and Non-Managers: Data • Use large cross sectional data sets for a group of high income countries. • 20 countries in our sample Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Iceland, Ireland, Israel, Italy, Luxembourg, Netherlands, Norway, Spain, Sweden, Switzerland, UK, US. • Divide individuals according to reported occupation. • Who is a manager?
Who is a Manager? US Chief executives and public administrators, Financial managers, Human resources and labor relations managers, Managers and Specialists in marketing, advertising, and public relations, Man. in educ. and related fields, Man. of medicine and health occupations, Postmasters and mail superintendents, Managers of food services and lodging occupations, Managers of properties and real estate, Funeral directors, Managers of service organizations, Managers and administrators
Age-Earnings Profiles Estimate age effects in earnings. We estimate
Age-Earnings Profiles Estimate age effects in earnings. We estimate a , y , e = α + β 0 a + β 1 a 2 + γ 1 y + γ 2 e i + ε i x i a , y , e , • x i a , y , e : log-earnings of individual i , age a in year y with education e . • a : age • y : year dummy. e : education dummy Estimate this equation for managers and non-managers separately. Focus on age-earnings profiles implied by β 0 and β 1 .
1.8 Figure 3. Age-earnings profiles in the US 1.7 US (managers) US (nonmanagers) 1.6 1.5 1.4 1.3 1.2 1.1 1 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age
1.8 Figure 1. Age-earnings profiles for managers 1.7 Canada US 1.6 Belgium 1.5 Spain 1.4 1.3 1.2 1.1 1 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age
1.6 Figure 2. Age-earnings profiles for non-managers 1.5 Canada US Belgium 1.4 Spain 1.3 1.2 1.1 1 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age
Relative Managerial Earnings Focus on ’slope’ of earnings growth of managers to non-managers to output per worker in prime-working years.
Relative Managerial Earnings Focus on ’slope’ of earnings growth of managers to non-managers to output per worker in prime-working years. WE FIND: Managerial income grows faster with age than non-managerial income in richer countries.
Relative Managerial Earnings Focus on ’slope’ of earnings growth of managers to non-managers to output per worker in prime-working years. WE FIND: Managerial income grows faster with age than non-managerial income in richer countries. Finding similar to Lagakos, Moll, Porzio, Qian and Schoellman (2014).
Relative Managerial Earnings Focus on ’slope’ of earnings growth of managers to non-managers to output per worker in prime-working years. WE FIND: Managerial income grows faster with age than non-managerial income in richer countries. Finding similar to Lagakos, Moll, Porzio, Qian and Schoellman (2014). Finding robust to consideration of different subsamples of our data, cohort instead of year effects and presence of education controls.
GDP per Worker and Relative Income Growth Sample: benchmark. Slope: 0.569. Corr: 0.489 Luxembou .4 Switzerland Log (Income Growth, 50−54/25−29) Germany US_Census .2 Canada France UK Belgium Israel Norway Italy Austria Australia Sweden Ireland Netherlands 0 Finland Iceland −.2 Spain Denmark 11 11.2 11.4 11.6 Log (GDP per Worker)
Model
Model • Life-cycle economy. j = 1, ... R , ... N . • Heterogenous individuals, all born with an initial endowment of managerial ability, z 1 ( t ) = G z ( t ) z . • z is drawn from a cumulative distribution F ( z ) • G z ( t ) grows at rate 1 + g z • Individuals can be workers or managers. Workers earn wage w . • Managers operate Lucas (1978) technology. They rent capital and labor services, produce output and collect managerial rents. • Occupational choice takes place at start of life, and it is irreversible.
Recommend
More recommend
