Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Upgrading and Shared Prosperity:
Some Insights from Micro Research
Eric Verhoogen
Columbia University
Brasilia, July 1, 2015
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Introduction
I There is broad agreement that innovation and productivity
improvements in manufacturing are a key ingredient in
economic growth.
I Important point from Mark’s talk: not all innovation is
new-to-the-world.
I All forms of upgrading and the firm level can contribute to
growth.
I Key question 1: How can we promote upgrading?
I Key question 2: What are the links between upgrading and
shared prosperity ?
I Within sectors, firms that do more upgrading also tend to pay
higher wages.
I Direction of causality unclear.
I Here I will talk about 5 insights into these questions from my
micro research on manufacturing firms in developing countries.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 1: Input quality matters for output quality
1
.5
0
−.5
−1
−1.5
log real output price, dev. from year means
1.5
A. Output prices, hollow brick (ladrillo hueco)
slope=−0.028, s.e.=0.032
−3
−2
−1
0
1
2
3
log employment, deviated from year means
0
−5
log real input price, dev. from year means
5
B. Input prices, common clay, paid by producers of hollow brick
slope=0.026, s.e.=0.073
−3
−2
−1
0
1
2
3
log employment, deviated from year means
I
Fig. 1 from Kugler & Verhoogen, REStud 2012.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 1: Input quality matters for output quality
−2
−1
0
1
2
3
log employment, deviated from year means
1.5
1
.5
0
−.5
log real input price, dev. from year means
1
.5
0
−.5
−3
−1.5 −1
1.5
B. Input prices, refined rendered suet, paid by producers of bar soap
slope=0.110, s.e.=0.038
−1.5 −1
log real output price, dev. from year means
A. Output prices, bar soap
slope=0.055, s.e.=0.025
−3
−2
−1
0
1
2
3
log employment, deviated from year means
1.5
1
.5
0
−.5
−1.5 −1
log real input price, dev. from year means
C. Input prices, unrefined rendered suet, paid by producers of bar soap
slope=0.103, s.e.=0.039
−3
−2
−1
0
1
2
3
log employment, deviated from year means
I
Fig. 2 from Kugler & Verhoogen, REStud 2012.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 1: Input quality matters for output quality
I On average, larger, more-productive firms use higher-quality
inputs, produce higher-quality outputs than smaller firms.
I Colombian manufacturing overall more like bar soap than like
hollow bricks.
I Industrial upgrading requires upgrading of entire complex of
final-good producers and input suppliers.
I Imported inputs tend to be higher-quality, may be important
for facilitating quality upgrading of final goods (Kugler &
Verhoogen, AER P&P, 2009).
I Related idea: producing new varieties of outputs may require
new varieties of inputs (Goldberg et al., 2010)
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 2: Exports ↑ → Quality ↑, Wages ↑
I New Beetle, almost all
exported.
e in
sco,
750
etles
City,
500
I Old Beetle, almost all sold
domestically (produced
until 2003).
I Source: Verhoogen, QJE
2008.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 2: Exports ↑ → Quality ↑, Wages ↑
Figure IV
Exports, High-quality Models as Percentage of VW Output
100
90
Percentage of total output
80
70
60
50
40
% output exported
30
% output New Beetle/Jetta/Golf
20
10
0
1988
1990
1992
1994
1996
1998
2000
2002
Notes: Output measured in physical units. Omitted model from upper curve is the Original Beetle. Data from Bulletins of
the Asociacion Mexicana de la Industria Automotriz (Mexican Automobile Industry Association).
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 2: Exports ↑ → Quality ↑, Wages ↑
I Técnico: 9 yrs. education,
2003 starting wage
∼US$11/day.
rs
[not
ars + 3
I Especialista: 12 yrs.
education, 2003 starting
wage ∼US$18/day.
ay
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 2: Exports ↑ → Quality ↑, Wages ↑
ions, Levels of Key Variables vs. log Domestic Sales, 1993 and 1997
App. Fig. IVb: Log white-collar wage
log real white-collar wage
0.75
94
97
0.5
0.25
0
1993
-0.25
1997
-0.5
-0.75
2
-3
-2
-1
0
log domestic sales 1993
1
2
I Non-parametric regressions, variables deviated from industry-year means.
App. Fig. IVd: Log wage ratio
I Similar patterns hold for ISO 9000 certification.
0.3
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 2: Exports ↑ → Quality ↑, Wages ↑
metric Regressions, Changes 1993-1997 and 1997-2001
App. Fig. Vb: Changes in log white-collar wage
Δ log real white-collar wage
0.2
7
1
0.1
0
-0.1
change 1993-1997
change 1997-2001
-0.2
-0.3
2
-3
-2
-1
0
log domestic sales, initial year
1
2
Fig. Vd: Changes
in log wage variables
ratio
IApp.
Non-parametric
regressions,
deviated from industry-year means.
I Similar patterns hold for ISO 9000 certification.
0.05
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 3: Exports ↑ → Wage premia ↑
−1
0
1
2
3
.3
.2
.1
0
−.1
avg. person component, deviated
−.3
−.2
.2
.1
0
−.3
−.2
−.1
plant component, deviated
.2
.1
0
−.1
−3
−2
−1
0
1
2
3
−3
−2
−1
0
1
2
1993−1997
1997−2001
−3
−2
−1
0
1
2
log domestic sales, initial year, deviated
3
.05
0
−.05
−.1
.05
0
−.05
.05
0
−.05
1993−1997
1993−1997
1997−2001
−3
−2
−1
0
1
2
3
log domestic sales, initial year, deviated
3
.1
Figure 9f. Changes in person component
change in avg. person component, deviated
Figure 9e. Changes in plant component
.1
Figure 9d. Changes in avg. log daily wage
change in plant component, deviated
log domestic sales, 1993, deviated
.1
log domestic sales, 1993, deviated
−.1
avg. log real daily wage, deviated
−.2
−.3
−2
log domestic sales, 1993, deviated
−.1
change in avg. log real daily wage, deviated
−3
I
Figure 9c. Person component
.3
Figure 9b. Plant component
.3
Figure 9a. Avg. log daily wage
1997−2001
−3
−2
−1
0
1
2
3
log domestic sales, initial year, deviated
Source: Frı́as, Kaplan & Verhoogen, Unpub. 2011.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 4: Adoption not Automatic
I Ongoing project with Atkin,
Chaudhry, Chaudry and Khandelwal,
2015.
I Setting: Soccer-ball cluster in Sialkot,
Pakistan
I ∼30 million balls/year, almost all
exported.
I 40% of world production, 70%
within hand-stitched segment
(WSJ, 2010).
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
1st Stage: Glue Cotton/Polyester to Artificial Leather
More on industry
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
2nd Stage: Cut Hexagons and Pentagons
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
3rd Stage: Print Logos/Designs on Panels
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
4th Stage: Stitch Panels around Bladder
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Existing Cutting Technology
Standard “buckyball” design:
20 hexagons, 12 pentagons.
For standard ball, almost all
firms use 2-hexagon and 2pentagon “flush” dies.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Existing Cutting Technology (cont.)
Hexagons tessellate. ∼ 8% of rexine wasted.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Existing Cutting Technology (cont.)
Pentagons don’t. ∼ 20-24% of rexine wasted.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Origin of Idea
In a YouTube video of a Chinese factory producing the Adidas
Jabulani ball, I noticed a different layout of pentagons.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
to Matters
one-half ofExports
the length
of d. These
twoPremia
chords define
a parallelogram
Input Quality
and Quality
Wage
Adoption
not Automaticq(d)Worker Buy-In Matters
inscribed in K, which turns out to be extensive (see Lemma 1 of the following
section). Now vary d and find a critical position of d = do such that q(do) is of
minimum area. This minimum-area extensive parallelogram generates a maximum
density double-lattice packing with copies of K. In general, locating the critical
diameter do may be a problem, but in many special cases, as in the following
examples, the diameter do is easy to find.
Origin of Idea (cont.)
We could also have gone to: G. Kuperberg and W. Kuperberg,
“Double-Lattice
Packings
oftheConvex
Bodiesin in
the 2Plane,”
Examples. An
application of
algorithm described
Remark
to the case Discrete
when K is a regular pentagon results in a double-lattice packing of density
& Computational
Geometry, 5: 389-397, 1990.
( 5 - x / 5 ) / 3 = 0 . 9 2 1 3 1 . . . , shown in Fig. 7. This packing may have the maximum
Fig. 7. Maximumdensity double-latticepacking with regular pentagons.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Origin of Idea (cont.)
Or the Wikipedia Pentagons page:
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Blueprint
         
    

       

  
Annalisa Guzzini (an architect, also
my wife)
and I developed
a
blueprint for a 4-pentagon die to implement the optimal packing.
39.5
54
5
I 44mm-edge
      pentagons:
             ∼250
     with
   old
    die
   vs.
  272
    with ours.
1-2
                              2     
I 43.5mm-edge pentagons: ∼258 vs. 280.
   X           272  5       39.5  54           

        
  
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
            
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
The “Shamyla” Die
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
20
0
10
cumulative number of dies
30
40
Die Purchases by Firm Z
Apr 12
Oct 12
Apr 13
Oct 13
Apr 14
date
I Second-largest by employment in Sialkot (∼2,200 employees).
I No-drop group, late responder.
I As of March 2014, using offset die for ∼100% of production.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Adoption as of Aug. 2013
Full sample
# ever active firms
# ever responded
# currently active and ever responded
# traded in
# ordered new die (beyond trade-in)
# received new die (beyond trade-in)
# ever used new die (>1000 balls, conservative)
# ever used new die (>1000 balls, liberal)
Upgrading and Shared Prosperity
Tech
Drop
Cash
Drop
No
Drop
Total
35
35
32
19
1
1
4
5
18
17
15
0
0
0
0
0
79
64
59
0
6
3
1
1
132
116
106
19
7
4
5
6
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 5: Worker Buy-In Matters
firm
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
no orders
to try on
too busy
2
2
2
2
2
4
3
3
3
1
1
1
3
3
2
1
5
2
doubt
profitable
waiting for
others to
prove value
waiting for
others to
iron out kinks
cutters
unwilling
3
3
other
production
issues
other
1
1
1
1
3
2
2
printing
problems
1
1
1
1
1
2
2
1
1
1
2
2
1
1
2
3
4
3
I Numbers indicate order of importance indicated by respondent.
I Sample is round-4 respondents who have had die in their factory but are not currently using it.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 5: Worker Buy-In Matters
I Why were cutters resisting?
I Most employees paid piece-rate and new technology slows
them down, at least initially.
I Cost savings accrue to owner.
I In absence of changes to labor contract, effective wage falls.
I A few owners changed labor contracts, but most did not.
I Either they simply did not realize that a change would be
desirable, or
I they found it too costly relative to expected benefits of
technology.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Insight 5: Worker Buy-In Matters
I We ran a second experiment in firms we had given technology
to:
I In some firms, we offered lump-sum bonus equal to 1 month
salary to cutter (and printer), if they could demonstrate
competence using the technology in presence of owner.
I Returned in one month to do test. All passed.
I Had significant effect on adoption:
I
I
Half of firms that could have responded ended up adopting.
None in control group in short term, one in medium term.
I One generalization we think we can draw: workers need to
expect to share in gains to adoption in order for adoption to
be successful.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Conclusion
I Some (tentative) lessons for Brazil:
I Quality upgrading is an important part of of innovation,
broadly defined, within firms.
I Upgrading of final goods requires access to high-quality inputs,
both foreign and domestic.
I Upgrading products and productivity also requires “upgrading”
the workforce:
I
I
In part this requires finding new higher-skilled workers.
But in larger part it requires motivating and training the
existing workforce.
I Labor relationships (and labor-market institutions) matter for
technology adoption/productivity improvement.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Conclusion
I Direction of causality between upgrading and shared
prosperity (i.e. wages, employment) still an open question.
I Standard view: innovations arrive exogenously, change skill
demand in firm.
I Alternative view: innovations arise endogenously, in part
through worker input. Workers have to have incentives to
share knowledge, ideas, good will.
I Results from Mexico, Pakistan provide some evidence for
alternative view. But there is still a lot of work to do.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
Conclusion
I This seems a promising direction for future work:
I Minimum wage study
I Evaluation of innovation-policy interventions.
I What are effects of innovation on wages, employment at firm
level?
I Is there an interaction of policies and labor-market conditions
in generating innovation?
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia
Intro
Input Quality Matters
Exports and Quality
Wage Premia
Adoption not Automatic
Worker Buy-In Matters
References I
Atkin, David, Azam Chaudhry, Shamyla Chaudry, Amit K. Khandelwal, and Eric Verhoogen, “Organizational
Barriers to Technology Adoption: Evidence from Soccer-Ball Producers in Pakistan,” 2015. Mimeo, Columbia
University.
Goldberg, Pinelopi K., Amit Khandelwal, Nina Pavcnik, and Petia Topalova, “Imported Intermediate Inputs and
Domestic Product Growth: Evidence from India,” Quarterly Journal of Economics, 2010, 125 (4), 1727–1767.
Kugler, Maurice and Eric Verhoogen, “Plants and Imported Inputs: New Facts and an Interpretation,” American
Economic Review Papers and Proceedings, May 2009, 99 (2), 501–507.
and
, “Prices, Plant Size and Product Quality,” Review of Economic Studies, January 2012, 79 (1),
307–339.
Kuperberg, G. and W. Kuperberg, “Double-Lattice Packings of Convex Bodies in the Plane,” Discrete &
Computational Geometry, 1990, 5, pp. 389–397.
Verhoogen, Eric, “Trade, Quality Upgrading and Wage Inequality in the Mexican Manufacturing Sector,” Quarterly
Journal of Economics, 2008, 123 (2), 489–530.
Wright, Tom, “Pakistan Defends Its Soccer Industry,” Wall Street Journal, 2010. April 26, 2010.
Upgrading and Shared Prosperity
Eric Verhoogen, Columbia