Firm Entry and Exit and Aggregate Growth

Firm Entry and Exit and Aggregate Growth Jose Asturias (Georgetown University, Qatar) Sewon Hur (University of Pittsburgh) Timothy Kehoe (UMN, Mpls Fed, NBER) Kim Ruhl (Penn State) University of Texas Austin October 2017 0 / 58

What drives aggregate productivity growth? Is productivity growth due to Continuing firms? Entry and exit of firms? 1 / 58

What drives aggregate productivity growth? Is productivity growth due to Continuing firms? Entry and exit of firms? Foster, Haltiwanger, and Krizan (2001) Net entry accounts for 25% of U.S. productivity growth Brandt, Van Biesebroeck, and Zhang (2012) Net entry accounts for 72% of Chinese productivity growth 1 / 58

Firm entry and aggregate growth: empirics How does firm entry and exit contribute to aggregate productivity growth? During periods of rapid GDP growth During periods of slow GDP growth 2 / 58

Firm entry and aggregate growth: empirics How does firm entry and exit contribute to aggregate productivity growth? During periods of rapid GDP growth During periods of slow GDP growth Use plant-level data from Chile and Korea Review literature that uses identical decomposition Net entry is more important in periods of rapid growth Average contribution, rapid growth: 47 percent Average contribution, slow growth: 22 percent 2 / 58

Firm entry and aggregate growth: model Construct a model of firm entry and exit Analytic expressions of FHK decomositions 3 / 58

Firm entry and aggregate growth: model Construct a model of firm entry and exit Analytic expressions of FHK decomositions Calibrate to the United States Create three distorted economies Reduce entry costs Reduce barriers to technology adoption Reduce fixed continuation costs Rapid growth with high contribution of entry and exit Driven by changes in relative productivities of entrants and exiters, as in data 3 / 58

Data 3 / 58

Plan Decompose aggregate productivity growth Terms related to entry and exit of firms Terms related to growth in continuing firms Follow Foster, Haltiwanger, and Krizan (2001) 4 / 58

Plan Decompose aggregate productivity growth Terms related to entry and exit of firms Terms related to growth in continuing firms Follow Foster, Haltiwanger, and Krizan (2001) Use manufacturing plant data from Chile and Korea Periods of rapid growth vs. slow growth Look within the same country at two windows Avoids cross-country differences Uses consistent datasets Review comparable studies in the literature 4 / 58

Defining industry productivity Productivity of industry i: log Z it = s et log z et e E it s et : gross output share of plant e in time t in industry i z et : TFP of plant e in time t in industry i Change in productivity (window defined by t 1, t ): log Z it = log Z it log Z i,t 1 5 / 58

Estimating plant productivity Plant e in industry i production function log y eit = log z eit + β i k log k eit + β i l log l eit + β i m log m eit Following Foster et al. (2001) β i j : average cost shares of input j in industry i Robust to alternative methods to estimate z, such as Woolridge-Levinsohn-Petrin 6 / 58

Productivity decomposition of industry growth log Z it = log Z NE it + log Z C it log Zit NE : change due to entering/exiting plants log Z C it : change due to continuing plants 7 / 58

Net entry log Z NE it = e N it s et (log z et log Z i,t 1 ) }{{} entering plants e X it s e,t 1 (log z e,t 1 log Z i,t 1 ) }{{} exiting plants N it and X it are sets of entering and exiting plants entering plants is positive if entrants have high productivity (compared to initial aggregate productivity) exiting plants is negative if exiting plants are unproductive 8 / 58

Continuing plants log Z C it = e C it s e,t 1 log z et }{{} within plant + e C it (log z et log Z i,t 1 ) s et }{{} reallocation C it is the set of continuing plants within plant is average within-plant productivity growth reallocation is positive if relatively productive plants expand market share 9 / 58

Productivity growth and aggregation At the industry-level we determine 1. Productivity change 2. Productivity change from entry/exit 3. Productivity change from continuing plants To aggregate, weight each of these three components by gross output of industry (using average of beginning and end of window) 10 / 58

Decomposing productivity growth: Chile and Korea How does the net entry term change in Chile and Korea? Look within the same country at two windows Avoids cross-country differences Uses consistent datasets 11 / 58

Real GDP per working-age person 400 slow growth (3.0%) 2009-2014 fast growth (4.3%) 2001-2006 Index (1985=100) 200 fast growth (6.1%) 1992-1997 Korea Chile fast growth (4.0%) 1995-1998 slow growth (2.7%) 2001-2006 100 1985 1990 1995 2000 2005 2010 2015 12 / 58

Plant-level manufacturing data Chile Encuesta Nacional Industrial Anual Collected by the Chilean national statistical agency Covers all plants with more than 10 employees 127 industries and 5,500 plants (2005) Panel: 1995 2006 Korea Survey of Mining and Manufacturing Collected by the Korean national statistical agency Covers all plants with more than 10 employees 180 industries and 68,640 plants (2014) Three panels: 1992 1997, 2001 2006, and 2009 2014 13 / 58

Net entry contribution higher in rapid growth Country Period GDP WAP Aggregate Effect of growth, productivity net entry annual growth, annual (percent) (percent) (percent) Chile 1995 1998 4.0 3.3 50.4* Chile 2001 2006 2.7 1.9 22.8 Korea 1992 1997 6.1 3.6 48.0 Korea 2001 2006 4.3 3.3 37.3 Korea 2009 2014 3.0 1.5 25.1 *: 5-year equivalent Results robust to other decompositions: Griliches and Regev (1995) and Melitz and Polanec (2015) details 14 / 58

Net entry, further decomposed Entering term can be decomposed into Market share of entrants (t) Average entrant productivity (t) relative to aggregate productivity (t 1) Exiting term can be decomposed into Market share of exiters (t 1) Average exiter productivity (t 1) relative to aggregate productivity (t 1) 15 / 58

FHK entering, decomposed multiplicatively Rapid growth features more-productive entrants Country Period FHK Relative Market entering productivity share of of entrants entrants (percent) (percent) Chile 1995 1998 6.6* 28.1* 0.24* Chile 2001 2006 2.5 6.8 0.36 Korea 1992 1997 5.6 15.0 0.38 Korea 2001 2006 2.0 7.3 0.27 Korea 2009 2014 0.6 2.4 0.27 *: 5-year equivalent 16 / 58

FHK exiting, decomposed multiplicatively Rapid growth features less-productive exiting plants Country Period FHK Relative Market exiting productivity share of of exiters exiters (percent) (percent) Chile 1995 1998 1.1* 5.7* 0.20* Chile 2001 2006 0.2 0.9 0.23 Korea 1992 1997 3.7 10.5 0.35 Korea 2001 2006 4.6 18.9 0.24 Korea 2009 2014 2.6 10.5 0.24 *: 5-year equivalent 17 / 58

Other empirical studies Existing studies with identical methodology Slow growth: Japan, Portugal, U.K., U.S. Rapid growth: Chile, China, Korea 18 / 58

Other empirical studies Existing studies with identical methodology Slow growth: Japan, Portugal, U.K., U.S. Rapid growth: Chile, China, Korea Problem: Studies use different length time windows Makes comparisons difficult Solution: Use calibrated model to make adjustments 18 / 58

Use model to make window adjustments Solve the baseline equilibrium for the U.S. Decompose model output using 5, 10, 15 year windows Fit a quadratic to contribution of net entry to productivity growth for the 3 windows 19 / 58

Net entry under various windows in the model Contribution of net entry to aggregate productivity 60 50 40 30 20 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Window length (years) 20 / 58

Use model to make window adjustments Portugal: 3-year window, 19 percent net entry contribution In the calibrated model 5 year window generates 25 percent contribution 3 year window generates 18 percent contribution Adjust proportionally Adjustment: 19 25/18 = 26 (5-year window equivalent) 21 / 58

Net entry more important during fast growth 60 CHN 1998-01* CHN 2001-07* Contribution of net entry (percent) 50 40 30 20 10 USA 1977-82 PRT 1991-94* KOR 2001-06 USA 1987-92 JPN 1994-01* KOR 1990-98* KOR 2009-14 CHL 2001-06 CHL 1995-98* KOR 1992-97 PRT 1994-97* USA 1982-87 GBR 1982-87 CHL 1990-97* 0-1 0 1 2 3 4 5 6 7 8 9 10 GDP (per 15-64) growth rate (percent) * denotes 5-year equivalents Average contribution: 47 percent (fast) and 22 percent (slow) 21 / 58

Model 21 / 58

Model We develop a model in which Potential entrants draw from frontier efficiency distribution, which improves by growth factor g e Efficiency of continuing firms grows Endogenous entry/exit of firms (and exogenous exit) Tractable balanced growth path BGP growth factor ge BGP level determined by barriers to entry, technology adoption Use model to investigate the relation between productivity growth and net entry 22 / 58

Household problem Representative household solves subject to max C t,b t+1 β t log C t t=0 P t C t + q t+1 B t+1 = w t + D t + B t C t 0, No Ponzi condition, B 0 given D t : aggregate dividends Normalize P t = 1, t 23 / 58

Firm dynamics Based on Hopenhayn (1992) Continuum of perfectly competitive firms A firm in the model is a plant in the data Heterogenous in efficiency x Productivity depends on efficiency Measure productivity as in data Pay κ to draw initial efficiency, f to operate Exogenous exit probability δ and endogenous exit 24 / 58

Fixed costs paid by firms Potential entrants pay κ t = κg t e to draw efficiency x κ = κ T (1 + τ) Paid in consumption/investment good g e : growth factor κ T : the technological cost, common across countries τ 0: policy induced barriers to entry 25 / 58

Firms face two decisions 1. Entry/exit decision 2. Conditional on operating: maximize profits 26 / 58

Firm s static problem Conditional on operating, firm with efficiency x solves d t (x) = max l t(x) xl t(x) α w t l t (x) f t Solution is l t (x) = ( ) 1 wt α 1 αx More efficient firms are larger 27 / 58

Firm s dynamic problem Firms with efficiency x choose to exit or continue to solve V t (x) = max {d t (x) + q t+1 (1 δ)v t+1 (g c,t+1 x), 0} Efficiency grows by g c,t+1 28 / 58

Operating firm efficiency growth Efficiency of existing firms grow by g ct = ḡg ε t ḡ is constant g t is average efficiency growth ε measures the degree of spillovers 29 / 58

Operating firm efficiency growth Efficiency of existing firms grow by g ct = ḡg ε t ḡ is constant g t is average efficiency growth ε measures the degree of spillovers Quantitatively, but not qualitatively important Further discussion in calibration 29 / 58

New entrant s problem Potential entrants draw efficiency from F t (x) = 1 ( ϕx g t e ) γ, x g t e ϕ Mean grows by growth factor g e Barrier to technology adoption, ϕ (Parente-Prescott 1994) Mass of potential entrants, µ t, from free entry condition: E x [V t (x)] = κ t Firm enters if and only if x ˆx t 30 / 58

Measure of firms Measure of firms of age j in operation η jt = µ t j+1 (1 δ) j 1 [ Convert age-j efficiency to initial efficiency g jt = Total measure of operating firms j 1 g c,t s+1 s=1 η t = η it i=1 ( )] ˆxjt 1 F t j+1 g jt 31 / 58

Equilibrium definition Given initial conditions, an equilibrium is Household consumption and bond plans Allocations and entry/exit thresholds for firms Measure of potential entrants for firms Prices and aggregate dividends 32 / 58

Equilibrium definition such that Household maximizes lifetime utility Firms maximize discounted dividends Costly entry condition binds Goods, labor, and bond markets clear Dividends satisfy [ ( )] D t = µ t j+1 (1 δ) j 1 ˆxt d t (x)df t j+1 µ t κ t j=1 ˆx jt g jt 33 / 58

Existence of balanced growth path Equilibrium converges to a balanced growth path in which 1. Entry and exit thresholds grow by g e 2. Real consumption, output, wages, and dividends grow by g e 3. Masses of potential entrants and operating firms are constant 34 / 58

Characterizing BGP: growth ˆx t = g e t ( ) 1 ωµ γ ϕ η w t = ( ) 1 α ( ) 1 α 1 α 1 α α ψ η ˆx t, Y t = ψ η ˆx t µ = ξ γλκω, η = ψ λf λ = ϕ 1 α κ 1 γ(1 α) 1 γα f γα ν Economy grows by g e ξ, ω, ν, ψ are constants λκ measures fixed cost, relative to output per capita fixed costs as labor 35 / 58

Comparative statics: entry costs µ = ξ γλκω, η = ψ λf ˆx t = g e t ( ) 1 ω ϕ η µ γ w t = ( ) 1 α ( ) 1 α 1 α 1 α α ψ η ˆx t, Y t = ψ η ˆx t λ = ϕ 1 α κ 1 As κ decreases γ(1 α) 1 γα f γα More potential entrants pay to draw efficiency (µ ) More-efficient firms enter (ˆxt ) Indirect effect: fixed costs relatively cheaper (λ ) ν Overall: wages and output increase 36 / 58

Comparative statics: barriers to adoption µ = ξ γλκω, η = ψ λf ˆx t = g e t ( ) 1 ω ϕ η µ γ w t = ( ) 1 α ( ) 1 α 1 α 1 α α ψ η ˆx t, Y t = ψ η ˆx t λ = ϕ 1 α κ 1 As ϕ decreases γ(1 α) 1 γα f γα Potential entrants draw higher efficiency (ˆxt ) Indirect effect: fixed costs relatively cheaper (λ ) ν Overall: wages and output increase 37 / 58

Comparative statics: continuation costs µ = ξ γλκω, η = ψ λf ˆx t = g e t ( ) 1 ω ϕ η µ γ w t = ( ) 1 α ( ) 1 α 1 α 1 α α ψ η ˆx t, Y t = ψ η ˆx t λ = ϕ 1 α κ 1 As f decreases γ(1 α) 1 γα f γα More firms operate (η ) Less-efficient firms produce (ˆxt ) Indirect effects: fixed costs relatively cheaper (λ ) ν Overall: wages and output increase 38 / 58

quantitative exercise 38 / 58

Removing distortions 1. Calibrate model to U.S. (high BGP) No policy distortions 2. Model three distorted economies on lower BGP Income level is 15 percent lower than U.S. κd = 1.94 κ us ϕd = 1.12 ϕ us f = 1.93 fus 3. Remove distortions Solve for transition to higher balanced growth path Analyze changes to productivity and contribution of net entry 39 / 58

Major reforms in Korea 1990 95 (affecting 1992 97 window) Relax FDI restrictions (1991,1992,1995) tech adoption Pro-competition reforms (1990,1993,1994) entry Financial reforms (1993,1994) entry & tech adoption 1998 2004 (affecting 2002 07 window) Abolish FDI restrictions (1998) tech adoption Pro-competition reforms (1999,2004) entry Financial reforms (1998,2004) entry & tech adoption Reforms to labor and corporate governance (1998 2001) Pace of reforms that affect productivity has slowed since 2004 40 / 58

Major reforms in Chile Reforms in the 1970s and early 1980s successful performance of the late 1980s and 1990s (Bergoeing et al. 2002) 1993 97 (affecting 1995 98 window) Relax FDI restrictions (1993) tech adoption Financial reforms (1993,1997) entry & tech adoption Privatization/deregulation of services (1993,1997) Pace of reforms that affect productivity has slowed since 1998 41 / 58

Measuring productivity Need model measurement consistent with data measurement Productivity z of firm with efficiency x log(z t (x)) = log(p t y t (x)) α lt log(l t (x)) α kt log(k t ) where α lt and α kt denote labor and capital shares Firm capital: k t = κ t + f t details Total factor productivity (our measure): log(z t (x)) = log(x) α kt log(κ t + f t ) Labor productivity (alternative measure): P ty t (x) l t (x) = w t α 42 / 58

FHK contribution of entry and exit In the balanced growth path, FHK contribution of entry: 1 (1 δ) ( gc exit: (1 δ) ( gc g e When g c = g e only exogenous exit g e ) γ 1 ) γ 1 α log(g e ) log(g c ) (1 α k ) log(g e ) FHK contribution of entry and exit: δ FHK contribution of entry and exit decreasing in g c conditions Distortions affect levels, but not FHK contributions 43 / 58

Calibration Calibrate model to match Size distribution of plants Effect of continuing plants on aggregate productivity growth Employment share of exiting plants 44 / 58

Calibrated parameters Model period is 5 years Data from United States Parameter Value Target Operating cost f T 0.46 5 average establishment size: 14.0 Entry cost κ T 0.38 entry cost / fixed cost : 0.82 Pareto parameter γ 6.10 establishment size s.d.: 89.0 Firm growth ḡ (1.006) 5 incumbent effect on growth : 75% Death rate δ 1 (0.96) 5 employment share (exiters): 19.3% Growth factor g e (1.02) 5 long-run growth: 2 percent Discount factor β (0.98) 5 real interest rate : 4 percent Returns to scale α 0.67 BGP labor share: 0.67 0.85 Barseghyan and DiCecio (2011); Foster et al. (2001); McGrattan and Prescott (2005) 45 / 58

Technological spillovers Take logs of equation that characterizes spillovers log g ct = log ḡ + ε log g t We estimate this equation as follows log g ct,i = β 0 + ε log g it + υ it gct,i is weighted productivity growth of continuing plants in i git is weighted productivity growth of entire industry i Estimate using Chile and Korea data (would like U.S. data) Average estimate: ε = 0.64 sensitivity 46 / 58

Output per worker 400 Real GDP (model period 0 = 100) 200 100 0 2 4 6 8 10 Model periods 47 / 58

Entry cost reform: more potential entrants More potential entrants increases efficiency thresholds Mass of potential entrants Detrended efficiency thresholds 0.6 Mass of potential entrants 0.5 0.4 0.3 0.2 0.1 Detrended efficiency thresholds, (model period 0 = 100) 115 110 105 100 0 2 4 6 8 10 Model periods 0 2 4 6 8 10 Model periods 48 / 58

Entry cost reform: more entry and exit Efficient firms enter, inefficient firms exit Mass of entering firms Mass of exiting firms Mass of successful entrants 0.03 0.02 0.01 Mass of exiting firms 0.03 0.02 0.01 0 2 4 6 8 10 Model periods 0 2 4 6 8 10 Model periods 49 / 58

Entry cost reform: wages and output More efficient firms increase wages and output Detrended wage Detrended output 120 120 115 115 110 110 105 105 100 0 2 4 6 8 10 model periods 100 0 2 4 6 8 10 model periods 50 / 58

Productivity growth decompositions Model Entry Annual output Contribution of periods cost growth net entry (5 years) (percent) (percent) 0 3 0.74 2.0 25.0 4 (reform) 0.38 4.6 59.6 5 0.38 2.5 36.5 6 0.38 2.1 28.1 7+ 0.38 2.0 25.0 labor as amalgam fixed costs denominated in labor 51 / 58

Remove barriers to technology adoption Smaller increase in mass of potential entrants Other figures are identical 0.6 0.5 Mass of potential entrants 0.4 0.3 0.2 0.1 technology adoption reform entry cost reform 0 2 4 6 8 10 Model periods 52 / 58

Productivity growth decompositions Model Barrier to Annual output Contribution of periods technology growth net entry (5 years) adoption (percent) (percent) 0 3 1.12 2.0 25.0 4 (reform) 1.00 4.6 59.6 5 1.00 2.5 36.6 6 1.00 2.1 28.1 7+ 1.00 2.0 25.0 53 / 58

Net entry and productivity in model and data Model generates quantitatively reasonable numbers Model Model Data Data reform BGP rapid slow GDP/WAP growth (percent) 4.6 2.0 5.8 2.1 Contribution of NE (percent) 60 25 47 22 54 / 58

FHK entering, decomposed multiplicatively Reforms increase relative productivity of entrants, as in data Model Model Data Data reform BGP rapid slow FHK entering (percent) 5.9 1.3 4.7 1.0 relative productivity (percent) 14.1 6.4 16.8 2.2 entrant market share 0.42 0.20 0.30 0.32 *: averages for Chile and Korea details 55 / 58

FHK exiting, decomposed multiplicatively Reforms decrease relative productivity of exiters, as in data Model Model Data Data reform BGP rapid slow FHK exiting (percent) 2.5 0.3 3.1 1.2 relative productivity (percent) 9.5 1.6 11.7 4.8 exiting plant market share 0.27 0.19 0.26 0.24 *: averages for Chile and Korea details 56 / 58

Not all reforms lead to productivity growth Lower continuation costs less productive firms operate Lowers aggregate productivity (more capital) Higher consumption and output Model Annual Ann. output Ann. prod. Contribution periods cont. growth growth of net entry (5 years) cost (percent) (percent) (percent) 0 3 0.90 2.0 1.3 25.0 4 0.46 4.1 1.1 88.8 5 0.46 2.8 1.1 36.9 6 0.46 2.2 1.2 28.5 7+ 0.46 2.0 1.3 25.1 57 / 58

Reform and growth Reforms that increase aggregate productivity feature Entry of more productive plants Exit of less productive plants Increase the contribution of net entry Need models of entry and exit to understand productivity growth 58 / 58

Thank you 58 / 58

Appendix 58 / 58

Defining industry productivity back Productivity of industry i: log Z it = s et log z et e E it s et : gross output share of plant e in time t in industry i z et : TFP of plant e in time t in industry i Change in productivity (window defined by t 1, t ): log Z it = log Z it log Z i,t 1 59 / 58

Estimating plant productivity back Plant e in industry i production function log y eit = log z eit + β i k log k eit + β i l log l eit + β i m log m eit Following Foster et al. (2001) β i j : average cost shares of input j in industry i Robust to alternative methods to estimate z, such as Woolridge-Levinsohn-Petrin 60 / 58

Productivity decomposition of industry growth back log Z it = log Z NE it + log Z C it log Zit NE : change due to entering/exiting plants log Z C it : change due to continuing plants 61 / 58

Net entry back log Z NE it = e N it s et (log z et log Z i,t 1 ) }{{} entering plants e X it s e,t 1 (log z e,t 1 log Z i,t 1 ) }{{} exiting plants N it and X it are sets of entering and exiting plants entering plants is positive if entrants have high productivity (compared to initial aggregate productivity) exiting plants is negative if exiting plants are unproductive 62 / 58

Continuing plants back log Z C it = e C it s e,t 1 log z et }{{} within plant + e C it (log z et log Z i,t 1 ) s et }{{} reallocation C it is the set of continuing plants within plant is average within-plant productivity growth reallocation is positive if relatively productive plants expand market share 63 / 58

Productivity growth and aggregation back At the industry-level we determine 1. Productivity change 2. Productivity change from entry/exit 3. Productivity change from continuing plants To aggregate, weight each of these three components by gross output of industry (using average of beginning and end of window) 64 / 58

Other decompositions back Net entry more important during periods of rapid growth Country Period FHK GR MP Chile 1995-1998 50.4* 23.5* 22.4* Chile 2001-2006 22.8 10.8-50.9 Korea 1992-1997 48.0 43.1 3.9 Korea 2001-2006 37.3 31.5-2.7 Korea 2009-2014 25.1 25.6-16.5 65 / 58

Net entry more important during fast growth Country Period GDP15 64 growth Effect of NE 5 year equivalent Japan 1994 2001 1.1 29 23 Portugal 1991 1994 0.5 19 26 Portugal 1994 1997 3.4 11 16 UK 1982 1987 3.3 12 12 US 1977 1982 0.4 25 25 US 1982 1987 3.7 14 14 US 1987 1992 1.6 35 35 Chile 2001 2006 2.7 23 23 Korea 2009 2014 3.0 25 25 Average 2.1 22 China 1998 2001 6.4 41 58 China 2001 2007 9.4 62 54 Chile 1990 1997 6.4 49 39 Korea 1990 1998 4.3 57 41 Chile 1995 1998 4.0 36 50 Korea 1992 1997 6.1 48 48 Korea 2001 2006 4.3 37 37 Average 5.8 47 Sources: U.S.: Foster et al. (2002); U.K.: Disney et al. (2005); Portugal: Carreira and Teixeira (2008); China: Brandt et al. (2012); Chile (1990 97): Bergoeing and Repetto (2006); Korea (1990 98) Ahn et al. (2005) 66 / 58

Measuring capital back Fixed costs (κ t, f t ) are investments How are they accounted for In the firm s accounts? In the national accounts? 67 / 58

Measuring capital back Fixed costs (κ t, f t ) are investments How are they accounted for In the firm s accounts? In the national accounts? Aggregate investment = µ t κ t + η t f t Depreciation is the sum of Capital of firms that die or exit κt of potential entrants that do not enter ft minus costs of upgrading capital for continuing firms 67 / 58

Measuring productivity back Need model measurement consistent with data measurement Productivity z of firm with efficiency x log(z t (x)) = log(y t (x)) α log(l t (x)) α kt log(k t ) Firm capital: k t = κ t + f t Capital share is given by α kt = R tk t Y t where R t = 1 q t 1 + δ kt 68 / 58

The Importance of entry and exit back Country Period Aggregate FHK FHK productivity entering exiting growth (percent) (percent) (percent) Chile 1995-1998 17.6* 6.6* -1.1* Chile 2001-2006 9.8 2.5 0.2 Korea 1992-1997 19.5 5.6-3.7 Korea 2001-2006 17.7 2.0-4.6 Korea 2009-2014 7.7-0.6-2.6 *: 5-year equivalent Korea 1992 1997: (5.6+3.7)/19.5 100 = 48.0 69 / 58

FHK entering, decomposed multiplicatively back Country Period FHK Relative Market entering productivity share of of entrants entrants (percent) (percent) Chile 1995-1998 6.6* 28.1* 0.24* Chile 2001-2006 2.5 6.8 0.36 Korea 1992-1997 5.6 15.0 0.38 Korea 2001-2006 2.0 7.3 0.27 Korea 2009-2014 -0.6-2.4 0.27 *: 5-year equivalent Rapid growth features more productive entrants 70 / 58

FHK exiting, decomposed multiplicatively back Country Period FHK Relative Market exiting productivity share of of exiters exiters (percent) (percent) Chile 1995-1998 -1.1* -5.7* 0.20* Chile 2001-2006 0.2 0.9 0.23 Korea 1992-1997 -3.7-10.5 0.35 Korea 2001-2006 -4.6-18.9 0.24 Korea 2009-2014 -2.6-10.5 0.24 *: 5-year equivalent Rapid growth features more unproductive exiting plants 71 / 58

Model with fixed costs as labor back Potential entrants pay κ t = w t κ to draw efficiency x Firms pay fixed cost of operating, f t = w t f, or exit Labor market: [ ( )] x 1 = µ t j+1 (1 δ) j 1 [l t (x) + f ] df t j+1 +µ t κ j=1 ˆx jt g jt Goods market: [ ( )] x C t = µ t j+1 (1 δ) j 1 xl t (x) α df t j+1 j=1 ˆx jt g jt Rental rate of capital: R t = 1 q t (1 δ kt ) w t+1 w t Measured productivity: log(z(x)) = log(x) α k log(κ + f ) 72 / 58

Characterizing BGP with fixed costs as labor back ˆx t = g e t ( ) 1 ωµ γ ϕ η ( ) 1 α 1 α w t = α ˆx t αf ( ) 1 α 1 α Y t = ψα ˆx t αf ξ µ = γκω ψ η = ψ = γ(1 α) 1 ψ γf γω (γ 1)ω + ξ 73 / 58

Major reforms in China back 1996-2005 (affecting 1985-2007 window) relax FDI restrictions (1998,2000-01) tech adoption lower trade barriers (1996-97,2001-02) tech adoption deregulation/privatization (2001-02) entry financial reforms (2000) entry & tech adoption bankruptcy reforms (1999) 74 / 58

Net entry contributions decreasing in g c back FHK contribution of entry in the BGP is unconditionally decreasing in g c FHK contribution of exit in the BGP is decreasing in g c if log ( ) ge g c g e < 1 α γ(1 α) 1 75 / 58

Calibrated with labor as amalgam back Model period is 5 years Data from United States Parameter Value Target Operating cost f T 0.25 5 average establishment size: 14.0 Entry cost κ T 0.20 entry cost / fixed cost : 0.82 Pareto parameter γ 13.42 establishment size s.d.: 89.0 Firm growth ḡ 1.006 5 incumbent effect on growth : 75% Death rate δ 1 0.96 5 employment share (exiters): 19.3% Growth factor g e 1.02 5 long-run growth: 2 percent Discount factor β 0.98 5 real interest rate : 4 percent Returns to scale α 0.85 Atkeson and Kehoe (2005) Barseghyan and DiCecio (2011); Foster et al. (2001); McGrattan and Prescott (2005) 76 / 58

Decomposition with labor as amalgam back Model Entry Annual output Productivity Contribution of periods cost growth growth net entry (5 years) (percent) (percent) (percent) 0-3 1.80 2.0 1.7 25.0 4 (reform) 0.20 5.0 4.8 74.0 5 0.20 2.3 1.6 33.5 6 0.20 2.1 1.7 26.6 7+ 0.20 2.0 1.7 25.0 77 / 58

Sensitivity back Table: Contribution of net entry (percent) BGP reform ε = 0.38 25.0 73.9 ε = 0.83 25.0 46.9 baseline 25.0 59.6 78 / 58

Calibration with fixed costs as labor back Model period is 5 years Data from United States Parameter Value Target Operating cost f T 0.69 5 average establishment size: 14.0 Entry cost κ T 0.57 entry cost / fixed cost : 0.82 Pareto parameter γ 6.10 establishment size s.d.: 89.0 Firm growth ḡ 1.006 5 incumbent effect on growth : 75% Death rate δ 1 0.96 5 variable labor share (exiters): 19.3% Growth factor g e 1.02 5 long-run growth: 2 percent Discount factor β 0.98 5 real interest rate : 4 percent Returns to scale α 0.67 BGP labor share: 0.67 Barseghyan and DiCecio (2011); Foster et al. (2001); McGrattan and Prescott (2005) 79 / 58

Decomposition with fixed costs as labor back Model Entry Annual output Productivity Contribution of periods cost growth growth net entry (5 years) (percent) (percent) (percent) 0-3 1.53 2.0 1.3 25.0 4 (reform) 0.57 4.7 4.3 56.5 5 0.57 2.4 1.3 34.0 6 0.57 2.2 1.3 28.0 7+ 0.57 2.0 1.3 24.9 80 / 58

Plant size distribution in Korea Average: 15.7 Standard Deviation: 136.2 Skewness: 99.9 Kurtosis: 13370.1 81 / 58