Lecture 3: Quantifying the Gains fromIndustrial Policy

Andres Rodrıguez-Clare

May 2019

The Textbook Case for Industrial Policy

O Q∗

D

MC(Q∗)

Quantity

Price

MC(Q)

Q

SMC

s

2 / 1

This Paper

1. Estimation of sector-level economies of scale (“SMC”)

I Use trade data to reveal effect of sector size on productivityI Construct IV using predictor of domestic demandI Clear evidence of scale effects: SE ∈ [0.07,0.25]

2. Evaluation of gains from industrial policy (“Triangle”)

I Focus on multi-sector Ricardian model with EESI Baseline average gains from IP = 0.61%I A bit lower than gains from optimal trade policy

3 / 1

Multi-Sector Gravity

I Many countries (i, j) and sectors (k)

I One factor: labor

I CES import demand:

xij,k ≡Xij,k∑lXlj,k

= c−θkij,k Pθkj,k

withcij,k =

τij,kwiAi,k

and

Pj,k = vk

(∑i

c−θkij,k

)−1/θk

4 / 1

Multi-Sector Gravity with EES

I Instead ofcij,k =

τij,kwiAi,k

we havecij,k =

τij,kwiAi,kL

γki,k

I Firms and consumers take Li,k as given

5 / 1

Optimal Policy

I Government of country i:

I Objective = Maximize welfare of representative agentI Tools = trade taxes/subsidies, production taxes/subsidies

I Optimal Industrial Policy:

I Pigouvian motiveI Production subsidies si,k = γk

I Optimal Trade Policy:

I TOT manipulationI Export taxes txij,k = 1

1+θk[if country i is “small” and cannot

affect relative prices in ROW]

6 / 1

Using Trade Data to Reveal Productivity

I Recall

xij,k ≡Xij,k∑lXlj,k

= c−θkij,k Pθkj,k

I From here we get

cij,k = x−1/θkij,k Pj,k

I We see thatx−1/θkij,k

is a “trade-revealed” (inverse) measure of productivity(modulo Pj,k)

I In paper: this idea is much more general

7 / 1

Estimation Equation

I Combinecij,k = x

−1/θkij,k Pj,k

withcij,k =

τij,kwiAi,kL

γki,k

to get

x−1/θkij,k Pj,k =

τij,kwiAi,kL

γki,k

I Rearrange and take logs to get

1

θklnxij,k = lnPj,k − lnwi + γk lnLi,k + ln(Ai,kτ

−1ij,k)

8 / 1

Estimation Equation

I So we have

1

θklnxij,k = lnPj,k − lnwi + γk lnLi,k + ln(Ai,kτ

−1ij,k)

I With fixed effects this can be rewritten as

1

θklnxij,k = δj,k + δij + γk lnLi,k + ln ηij,k

I In paper: non-parametric identification

9 / 1

Empirical Strategy

1

θklnxij,k = δj,k + δij + γk lnLi,k + ln ηij,k

I Step 1: Construct LHS using TEs (θk) from literature

I Step 2: Construct demand-side IV for RHS, {lnZi,k}I If Cobb-Douglas, then use Zi,k = xj,k · LjI Generalize to CES

I Step 3: Estimate SEs (γk) using {lnZi,k} as IV for {lnLi,k}

10 / 1

Data

I OECD Inter-Country Input-Output tablesI 61 countriesI 34 sectors (15 manufacturing)I Focus on manufacturingI Years 1995, 2000, 2005, 2010

I Population from PWT v9.0

11 / 1

Sector-Level SEs (γk), Part I

Reduced- First-stage SWOLS IV form F-stat F-stat

Sector (1) (2) (3) (4) (5)

Food, Beverages and Tobacco 0.19 0.16 0.10 87.20 394.3(0.01) (0.02) (0.02)

Textiles 0.14 0.12 0.06 56.70 349.9(0.01) (0.01) (0.02)

Wood Products 0.13 0.11 0.05 15.50 210.7(0.01) (0.02) (0.01)

Paper Products 0.14 0.11 0.05 55.60 661.9(0.01) (0.02) (0.01)

Coke/Petroleum Products 0.09 0.07 0.03 14.20 299.1(0.01) (0.01) (0.01)

Chemicals 0.23 0.20 0.17 31.10 335.8(0.01) (0.02) (0.02)

Rubber and Plastics 0.29 0.25 0.22 39.13 436.0(0.02) (0.03) (0.03)

12 / 1

Sector-Level SEs (γk), Part II

Reduced- First-stage SWSector OLS IV form F-stat F-stat

Mineral Products 0.16 0.13 0.08 40.50 405.0(0.01) (0.02) (0.01)

Basic Metals 0.13 0.11 0.07 14.40 254.0(0.01) (0.01) (0.01)

Fabricated Metals 0.16 0.13 0.07 57.10 421.1(0.01) (0.02) (0.01)

Machinery and Equipment 0.15 0.13 0.07 66.40 401.6(0.01) (0.01) (0.01)

Computers and Electronics 0.10 0.09 0.04 18.60 290.5(0.01) (0.01) (0.01)

Electrical Machinery, NEC 0.11 0.09 0.03 45.90 419.5(0.01) (0.01) (0.01)

Motor Vehicles 0.17 0.15 0.15 39.80 390.2(0.01) (0.01) (0.02)

Other Transport Equipment 0.17 0.16 0.11 24.00 381.6(0.01) (0.02) (0.02) 13 / 1

Estimates of Welfare Gains

Table 3: Gains from Optimal Policies, Selected Countries

Optimal Ind. Policy Trade Policy Gains from Gains fromPolicy Only Only Trade Policy Ind. Policy

Country (1) (2) (3) (4) (5)

United States 0.65% 0.29% 0.28% 0.36% 0.37%

China 0.74% 0.42% 0.26% 0.32% 0.48%

Germany 1.11% 0.13% 0.49% 0.98% 0.62%

Ireland 2.32 % -0.41% 1.23% 2.73% 1.10%

Vietnam 1.74% 0.67% 1.07% 1.07% 0.67%

Avg., Unweighted 1.42% 0.41% 0.82% 1.01% 0.60%Avg., GDP-weighted 0.87% 0.30% 0.42% 0.58% 0.46%

14 / 1

The Textbook Case for Industrial Policy

O Q∗

D

MC(Q∗)

Quantity

Price

MC(Q)

Q

SMC

s

15 / 1

Explaining the Size of Gains

I Closed economy with outside good ⇒ gains approximately

∆W

Y=

1

2

∑k∈K

(LkL

)γ2k

1/ρ− γk

I Gains reflect size of subsidies (γ2k), and quantity response

to subsidies (LkL ·

11/ρ−γk )

I Using averages to approximate magnitude of gains,

∆W

Y=

1

2× (0.28)× (0.13)2

(1/1.5)− 0.13' 0.44%

16 / 1

Explaining the Pattern of Gains

I Closed economy with outside good ⇒ gains approximately

∆W

Y=

1

2

∑k∈K

(LkL

)γ2k

1/ρ− γk

I Given γk, in autarky gains limited by domestic demand

I Open economy faces more elastic international demandI Gains tend to be higher for more open countriesI Openness in high-γk AND low-γk importantI World economy is closed, so global gains constrained by 1/ρ

17 / 1

Gains from Industrial Policy Increase with Openness

ARG

AUS

AUTBEL BGR

BRA

BRN

CAN

CHE

CHL

CHNCOL

CRICYP

CZE

DEUDNK

ESP

EST

FINFRAGBR

GRC HKG

HRV

HUN

IDN

IND

IRL

ISL

ISR

ITA

JPN

KHMKOR

LTU

LUX

LVA

MEXMLT

MYS

NLD

NOR

NZL PHL

POL

PRTROURUS

SAU

SGP

SVKSVN

SWE

THA TUN

TUR

TWN

USA

VNM

ZAF

.4.6

.81

1.2

Gai

ns fr

om In

dust

rial P

olic

y

10 20 30 40 50Openness

18 / 1

Robustness: Non-Manufacturing γ, ρ

.1.3

.5.7

.91.

1A

vg. G

ains

from

Ind.

Pol

icy

0 .05 .1 .15 .2 .25Non - Manufacturing γ

3(a): Gains for Different Non-Manufacturing γ

.1.3

.5.7

.91.

1A

vg. G

ains

from

Ind.

Pol

icy

.5 1 1.5 2 2.5 3ρ

3(b): Gains for Different ρ

19 / 1

Robustness: Trade Elasticities

Table 5: Gains from Industrial Policy, Alternative Trade Elasticities

TruncatedBaseline Shapiro BSY CP GYY CP

Country (1) (2) (3) (4) (5) (6)

United States 0.37% 0.52% 0.32% 3.67% 1.33% 1.40%

China 0.48% 0.76% 0.42% 4.23% 1.54% 2.21%

Germany 0.64% 0.92% 0.52% 7.35% 1.90% 3.11%

Ireland 1.10 % 1.83% 0.77% 5.16% 3.09% 2.14%

Vietnam 0.69% 1.13% 0.56% 7.92% 1.93% 2.41%

Avg., Unweighted 0.60% 0.81% 0.52% 4.78% 1.84% 2.04%Avg., GDP-Weighted 0.46% 0.67% 0.39% 4.25% 1.48% 1.85%Corr. with Baseline - 0.60 0.84 0.18 0.74 0.29

20 / 1

Estimates of Welfare Gains

Table 4: Gains from Constrained and Globally Efficient IndustrialPolicies, Selected Countries

Baseline Constrained Globally EfficientIndustrial Policy Industrial Policy Industrial Policy

Country (1) (2) (3)

United States 0.37% 0.35% 0.42%

China 0.48% 0.45% 0.21%

Germany 0.62% 0.48% -0.35%

Ireland 1.10% 1.55% -1.78%

Vietnam 0.67% 0.99% 1.36%

Avg., Unweighted 0.60% 0.79% 0.29%Avg., GDP-Weighted 0.46% 0.48% 0.22%

21 / 1

Conclusion

I Skepticism about textbook IP: “we don’t know the SEs!”

I This paper shows how to estimate SEs using commonlyavailable trade and production data and then how tocompute gains from IP

I Evidence of SEs, with large variability across sectors

I Even an enlightened and benevolent social planner wouldachieve gains on the order of 1%

I A bit lower than gains from optimal trade policy

22 / 1