Organizational Barriers to Technology Adoption: Evidence from - - PowerPoint PPT Presentation

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Organizational Barriers to Technology Adoption: Evidence from Soccer-Ball Producers in Pakistan

David Atkin Azam Chaudhry Shamyla Chaudry

UCLA Lahore School

• f Economics

Lahore School

• f Economics

Amit Khandelwal Eric Verhoogen

Columbia University Columbia University

• Feb. 2015

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Introduction

◮ Research question: When firms are exposed to a beneficial technology, do they adopt?

◮ Will talk about what we mean by “beneficial.”

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Introduction

◮ Research question: When firms are exposed to a beneficial technology, do they adopt?

◮ Will talk about what we mean by “beneficial.”

◮ We invented a new technology and randomly allocated it to soccer-ball producers in Sialkot, Pakistan. (Details coming.)

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Introduction

◮ Research question: When firms are exposed to a beneficial technology, do they adopt?

◮ Will talk about what we mean by “beneficial.”

◮ We invented a new technology and randomly allocated it to soccer-ball producers in Sialkot, Pakistan. (Details coming.)

◮ Initial (and future) objective: study spillovers.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Introduction

◮ Research question: When firms are exposed to a beneficial technology, do they adopt?

◮ Will talk about what we mean by “beneficial.”

◮ We invented a new technology and randomly allocated it to soccer-ball producers in Sialkot, Pakistan. (Details coming.)

◮ Initial (and future) objective: study spillovers. ◮ Early take-up puzzlingly low.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Introduction

◮ Research question: When firms are exposed to a beneficial technology, do they adopt?

◮ Will talk about what we mean by “beneficial.”

◮ We invented a new technology and randomly allocated it to soccer-ball producers in Sialkot, Pakistan. (Details coming.)

◮ Initial (and future) objective: study spillovers. ◮ Early take-up puzzlingly low. ◮ This paper: how organizational barriers (agency problems) can

prevent adoption.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Setting: Soccer-Ball Cluster in Sialkot, Pakistan

◮ 135 firms (initially) ◮ ∼30 million balls/year, almost all exported. ◮ 40% of world production, 70% within hand-stitched segment (WSJ, 2010). ◮ Common technology producing standardized product.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

1st Stage: Glue Cotton/Polyester to Artificial Leather

More on industry Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

2nd Stage: Cut Hexagons and Pentagons

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

3rd Stage: Print Logos/Designs on Panels

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

4th Stage: Stitch Panels around Bladder

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Existing Cutting Technology

Standard “buckyball” design: 20 hexagons, 12 pentagons. For standard ball, almost all firms use 2-hexagon and 2- pentagon “flush” dies.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Existing Cutting Technology (cont.)

Hexagons tessellate. ∼ 8% of rexine wasted.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Existing Cutting Technology (cont.)

Pentagons don’t. ∼ 20-24% of rexine wasted.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Origin of Idea

In a YouTube video of a Chinese factory producing the Adidas Jabulani ball, I noticed a different layout of pentagons.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Origin of Idea (cont.)

We could also have gone to: G. Kuperberg and W. Kuperberg, “Double-Lattice Packings of Convex Bodies in the Plane,” Discrete & Computational Geometry, 5: 389-397, 1990.

Double-Lattice Packings

• f Convex Bodies

in the Plane 393 inscribed in Ko, and the proof is complete. However, it can be noticed now that the minimality of the area of q implies that the length of one of the sides of q actually equals one-half of the length of Ko in the direction of that side. Therefore Ko actually touches a translate of itself, and Case II is not possible at all. [] Remark 1. If K is not strictly convex, the conclusion of the above theorem does not necessarily hold. However, in this case there exists a double-lattice packing with maximum density which is generated by a minimum-area extensive parallelogram inscribed in K. This can be obtained by approximating K with a sequence of strictly convex bodies K, and then selecting a convergent subsequence

• f double-lattice packings.

Remark 2. Theorem 1 and the above remark yield an algorithm for finding a maximum density double-lattice packing with copies of K which goes as follows. For any diameter d of K, find a pair of chords parallel to d, each of length equal to one-half of the length of d. These two chords define a parallelogram q(d) 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. Examples. An application of the algorithm described in Remark 2 to the case when K is a regular pentagon results in a double-lattice packing of density (5-x/5)/3 =0.92131..., shown in Fig. 7. This packing may have the maximum

• Fig. 7. Maximum

density double-lattice packing with regular pentagons. Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Origin of Idea (cont.)

Or the Wikipedia Pentagons page:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Blueprint

Annalisa Guzzini (an architect, my wife) and I developed a blueprint for a 4-pentagon die to implement the optimal packing.

• ◮ 44mm-edge pentagons: ∼250 with old die vs. 272 with ours.

◮ 43.5mm-edge pentagons: ∼258 vs. 280.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Blueprint (cont.)

Blueprint includes instructions for modifying size of die.

• ◮ Sides of adjacent pentagons are “offset,” not flush.

◮ 4-pentagon pattern can be replicated by two 2-pentagon cuts.

◮ We will also consider a two-piece offset die as our technology

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

The “Shamyla” Die

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Net Variable Cost Reduction from Offset Die

◮ Values for median firm (see Table 2 for distribution). ◮ Cost reduction from reduced rexine wastage:

Cost reduction per pentagon: 7.9% Pentagons as share of rexine sheet cost: 33% Rexine sheet cost as share of total cost: 44.7% Estimated total cost reduction (7.9% × 33% × 44.7%): 1.12%

◮ Additional labor costs:

Increase in pentagon cutting time (conservative): 50% Pentagons as share of cutting time: 33% Cutting as share of total costs .45% Estimated total cost increase (50% × 33% × .45%): ∼.1%

◮ Net variable cost reduction: ∼1%.

◮ Small, but 12.3% of profits

Details Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Net Benefits of Adoption (Table 3)

10th 25th 50th 75th 90th mean net variable cost reduction (%) 0.52 0.72 1.02 1.29 1.87 1.09 (0.05) (0.03) (0.03) (0.06) (0.17) (0.04) % net variable cost/avg % profit rate 5.27 8.10 12.34 19.86 28.98 15.45 (0.42) (0.50) (0.73) (1.21) (2.26) (0.71) total cost savings per month (Rs 000s) 4.46 12.19 49.38 165.21 475.01 174.12 (0.60) (1.24) (5.43) (18.42) (79.89) (18.54) days to recover fixed costs 8.48 15.98 36.61 80.34 193.92 136.94 (0.97) (1.68) (2.93) (6.51) (15.64) (44.54) days to recover fixed costs (no die) 4.40 8.30 19.01 41.71 100.69 71.10 (0.51) (0.87) (1.52) (3.38) (8.12) (23.13)

◮ We estimate that 50% of tech drop firms would recover fixed costs in 19 days or less, 75% in 42 days or less. ◮ Possibility of unobserved fixed costs considered in paper.

Full table

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Tech Drop Experiment: Design

3 groups:

• 1. Tech drop:

◮ Die + blueprint. ◮ 30 min. demonstration, including comparison to existing die. ◮ Offer to trade in die for different size at no cost.

◮ Panel sizes vary, even for a given size ball. ◮ To be usable, pentagon die has to be exactly same size as

hexagon die.

• 2. Cash drop:

◮ 30,000 Rs cash (∼ US$300) — the amount we paid for each

die.

• 3. “No drop”

◮ No intervention.

Dropped technology in May-June 2012. Surveys approx. every 3 months since then.

Timeline Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Adoption as of Aug. 2013 (Table 7)

Tech Drop Cash Drop No Drop Total Full sample # ever active firms 35 18 79 132 # ever responded 35 17 64 116 # currently active and ever responded 32 15 59 106 # traded in 19 19 # ordered new die (beyond trade-in) 1 6 7 # received new die (beyond trade-in) 1 4 5 # ever used new die (>1000 balls) 5 1 6 # currently using new die (>1000 balls) 5 1 6

Full table Characteristics of exiters Adoption as of March 2014 Hypotheses for non-adoption Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Die Purchases by Firm Z

10 20 30 40 cumulative number of dies Apr 12 Oct 12 Apr 13 Oct 13 Apr 14 date

◮ Second-largest by employment in Sialkot (∼2,200 employees). ◮ No-drop group, late responder. ◮ As of March 2014, using offset die for ∼100% of production.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Table 10: Reasons for Non-Adoption

firm no orders to try on too busy doubt profitable waiting for

• thers to

prove value waiting for

• thers to

iron out kinks cutters unwilling printing problems

• ther

production issues

• ther

1 2 3 1 2 2 1 3 2 1 4 2 1 5 2 1 6 4 3 1 2 7 3 2 1 8 3 1 2 9 3 2 1 10 1 11 1 12 1 13 3 1 2 14 3 1 2 15 2 1 3 16 1 17 5 3 1 2 4 18 2 3 1 3

◮ Numbers indicate order of importance indicated by respondent. ◮ Sample is round-4 respondents who have had die in their factory but are not currently using it.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Die Purchases by Firm Z Redux

10 20 30 40 cumulative number of dies Apr 12 Oct 12 Apr 13 Oct 13 Apr 14 date

◮ Second-largest by employment in Sialkot (∼2,200 employees). ◮ No drop group, late responder. ◮ As of March 2014, using offset die for ∼100% of production. ◮ Pays monthly salary to cutters, not piece rate.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Brief Summary of Model

◮ Principal-agent model with non-contractible effort, limited liability for agent.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Brief Summary of Model

◮ Principal-agent model with non-contractible effort, limited liability for agent. ◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Brief Summary of Model

◮ Principal-agent model with non-contractible effort, limited liability for agent. ◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal

◮ Technology may be of 3 types, one of which resembles ours (material-saving, labor-using).

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Brief Summary of Model

◮ Principal-agent model with non-contractible effort, limited liability for agent. ◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal

◮ Technology may be of 3 types, one of which resembles ours (material-saving, labor-using). ◮ We show that there exist conditions under which:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Brief Summary of Model

◮ Principal-agent model with non-contractible effort, limited liability for agent. ◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal

◮ Technology may be of 3 types, one of which resembles ours (material-saving, labor-using). ◮ We show that there exist conditions under which:

◮ If only standard linear contract (that cannot be conditioned on

ex-post-revealed marginal cost) available, agent misinforms principal and a beneficial technology is not adopted.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Brief Summary of Model

◮ Principal-agent model with non-contractible effort, limited liability for agent. ◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal

◮ Technology may be of 3 types, one of which resembles ours (material-saving, labor-using). ◮ We show that there exist conditions under which:

◮ If only standard linear contract (that cannot be conditioned on

ex-post-revealed marginal cost) available, agent misinforms principal and a beneficial technology is not adopted.

◮ If contract can be conditioned on marginal cost at transaction

cost G, and G is sufficiently low relative to expected gains (which depend on principal’s prior), adoption occurs.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Incentive-Payment Experiment: Design

Randomly assign still-active tech-drop firms to:

• A. Incentive group:

◮ Refresher about technology. Offer repeat of demonstration.

Mention 2-pentagon die.

◮ Incentive treatment:

◮ Explain misaligned incentives to owner. ◮ Offer incentive payment to one cutter, one printer (US$150 or

US$120, roughly monthly income) if they can demonstrate competence using new technology.

◮ Pay 1/3 up front, 2/3 conditional on satisfactory performance

(272 pentagons in 3 min. for cutter, 48 2-pentagon swipes in 3 min. for printer) in 4-6 weeks.

◮ 20 rexine sheets to practice with. US$50 to owner to defray

• verhead costs (electricity, additional practice rexine).
• B. No-incentive group:

◮ Refresher about technology, offer repeat of demonstration,

mention 2-pentagon die.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Covariate Balance, Incentive Payment Experiment

Group A Incentive Payment Group B No Incentive Payment log avg output/month 9.86 9.31 (0.41) (0.29) log avg employment 3.35 3.23 (0.38) (0.25) log avg price, size 5 promo 5.40 5.45 (0.02) (0.07) log avg price, size 5 training 6.00 5.93 (0.06) (0.06) avg % promotional (of size 5) 34.90 32.04 (6.20) (7.26) avg Rs/ball, head cutter 1.45 1.63 (0.10) (0.15) CEO university indicator 0.56 0.36 (0.18) (0.15) CEO experience 15.50 16.50 (3.60) (3.60) age of firm 24.53 20.60 (2.83) (2.28) N 15 16

◮ No differences significant at 5% level. ◮ As of Aug. 2013, we believed there were 34 active tech drop firms, and randomized 17 firms each into groups A and B. Three were subsequently revealed to have closed and/or stopped producing soccer balls.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Incentive Payment Experiment: Summary

◮ Of the 15 active Group A (Incentive Payment) firms:

◮ 5 refused intervention ◮ 10 accepted (2 were already adopters)

◮ All passed, average time 2:52 (2:15 Traditional) “Test” results Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Incentive Payment Experiment: Summary

◮ Of the 15 active Group A (Incentive Payment) firms:

◮ 5 refused intervention ◮ 10 accepted (2 were already adopters)

◮ All passed, average time 2:52 (2:15 Traditional)

◮ Of the 8 accepters who had not yet adopted:

“Test” results Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Incentive Payment Experiment: Summary

◮ Of the 15 active Group A (Incentive Payment) firms:

◮ 5 refused intervention ◮ 10 accepted (2 were already adopters)

◮ All passed, average time 2:52 (2:15 Traditional)

◮ Of the 8 accepters who had not yet adopted:

◮ 5 have adopted (> 1000 balls). “Test” results Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Incentive Payment Experiment: Summary

◮ Of the 15 active Group A (Incentive Payment) firms:

◮ 5 refused intervention ◮ 10 accepted (2 were already adopters)

◮ All passed, average time 2:52 (2:15 Traditional)

◮ Of the 8 accepters who had not yet adopted:

◮ 5 have adopted (> 1000 balls). ◮ 3 (not strict subset) have purchased their first offset die

(beyond trade-in).

“Test” results Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Incentive Payment Experiment: Summary

◮ Of the 15 active Group A (Incentive Payment) firms:

◮ 5 refused intervention ◮ 10 accepted (2 were already adopters)

◮ All passed, average time 2:52 (2:15 Traditional)

◮ Of the 8 accepters who had not yet adopted:

◮ 5 have adopted (> 1000 balls). ◮ 3 (not strict subset) have purchased their first offset die

(beyond trade-in).

◮ Of the 16 active Group B (No Incentive) firms:

◮ 3 were already adopters. ◮ Of 13 non-adopters, none have adopted or purchased die.

“Test” results Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Table 13: Current Use as Outcome

• Dep. var.: currently using offset die (have produced > 1000 balls with it)

All Strata Initial Non-Adopters First Stage OLS Reduced Form (ITT) IV (TOT) First Stage OLS Reduced Form (ITT) IV (TOT) (1) (2) (3) (4) (5) (6) (7) (8) rec’d treatment 0.48*** 0.48*** 0.59*** 0.63*** (0.15) (0.15) (0.18) (0.18) assigned to group A 0.68*** 0.32** 0.62*** 0.38*** (0.12) (0.12) (0.14) (0.13) stratum dummies Y Y Y Y Y Y Y Y R-squared 0.57 0.66 0.59 0.66 0.50 0.43 0.26 0.43 N 31 31 31 31 26 26 26 26

◮ We also conduct small-sample-robust two-tailed permutation test. ◮ Calculate ITT estimate for all 25.8 million treatment assignments. ◮ Reject null if our estimate lies outside 2.5th, 97.5th percentile. ◮ p-value (corresponding to Column (3)) = .03

Permutation test coeefficient distribution Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Conclusion

◮ In paper we argue against alternative stories:

◮ We just paid fixed costs of adoption for firms. ◮ Second experiment just increased salience of technology.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Conclusion

◮ In paper we argue against alternative stories:

◮ We just paid fixed costs of adoption for firms. ◮ Second experiment just increased salience of technology.

◮ Results still preliminary. We are finishing up additional survey round.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Conclusion

◮ In paper we argue against alternative stories:

◮ We just paid fixed costs of adoption for firms. ◮ Second experiment just increased salience of technology.

◮ Results still preliminary. We are finishing up additional survey round. ◮ But results suggest that piece-rate-induced worker resistance is an important barrier to adoption.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Conclusion

◮ In paper we argue against alternative stories:

◮ We just paid fixed costs of adoption for firms. ◮ Second experiment just increased salience of technology.

◮ Results still preliminary. We are finishing up additional survey round. ◮ But results suggest that piece-rate-induced worker resistance is an important barrier to adoption.

◮ A relatively small intervention in monetary terms has had a

reasonably large impact on adoption.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Conclusion

◮ In paper we argue against alternative stories:

◮ We just paid fixed costs of adoption for firms. ◮ Second experiment just increased salience of technology.

◮ Results still preliminary. We are finishing up additional survey round. ◮ But results suggest that piece-rate-induced worker resistance is an important barrier to adoption.

◮ A relatively small intervention in monetary terms has had a

reasonably large impact on adoption.

◮ Consistent with explanation that workers were misinforming

• wners, and that intervention worked by inducing truthful

revelation.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Conclusion (cont.)

◮ Big picture:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Conclusion (cont.)

◮ Big picture:

◮ Inertia in labor contracts can hinder technological change.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Conclusion (cont.)

◮ Big picture:

◮ Inertia in labor contracts can hinder technological change. ◮ Workers need to expect to share in gains to adoption in order

for adoption to be successful.

◮ One answer to why “inclusiveness” matters for growth. Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

References I

Kuperberg, G. and W. Kuperberg, “Double-Lattice Packings of Convex Bodies in the Plane,” Discrete & Computational Geometry, 1990, 5, 389–397. Wright, Tom, “Pakistan Defends Its Soccer Industry,” Wall Street Journal, 2010. April 26, 2010. Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Setting: Soccer-Ball Cluster in Sialkot, Pakistan

◮ 70% of world production

• f hand-stitched balls

(WSJ, 2010). ◮ 40% of total world production. ◮ 135 active firms ◮ ∼30 million balls/year.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Industry Structure

◮ 5-10 large firms (250+ employees):

◮ Produce own brands or high-quality name-brand balls for

Adidas (including 2014 World Cup ball), Nike etc.

◮ Tend to be the innovators in sector.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Industry Structure

◮ 5-10 large firms (250+ employees):

◮ Produce own brands or high-quality name-brand balls for

Adidas (including 2014 World Cup ball), Nike etc.

◮ Tend to be the innovators in sector.

◮ Fringe of small- and medium-sized firms:

◮ Find clients at large industry expos, or subcontract for local

firms.

◮ Produce low-quality “promotional” balls, often with advertiser

logo, or medium-quality “training” balls, often with club logo.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Industry Structure

◮ 5-10 large firms (250+ employees):

◮ Produce own brands or high-quality name-brand balls for

Adidas (including 2014 World Cup ball), Nike etc.

◮ Tend to be the innovators in sector.

◮ Fringe of small- and medium-sized firms:

◮ Find clients at large industry expos, or subcontract for local

firms.

◮ Produce low-quality “promotional” balls, often with advertiser

logo, or medium-quality “training” balls, often with club logo.

◮ Almost all balls exported.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

• Fig. 2: U.S. Imports of Soccer Balls

10 20 30 40 50 total imports (f.o.b.) into US (mil. 2000 US$) 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 year

Pakistan China Other

◮ 10-digit HS category 9506.62.40.80 (inflatable soccer balls).

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Table 1: Pentagons/sheet

traditional die

• ffset die
• wner

report direct

• bs.
• wner

report direct

• bs.

(1) (2) (3) (4) size 43.5 257.4 257.7 273.5 277.5 (10.4) (6.7) (4.4) (5.3) size 43.75 256.3 254.4 269.0 272.0 (6.2) (9.4) (1.4) (0.0) size 44 253.8 248.4 280.0 272.5 (8.4) (18.7) (0.7) size 44.25 246.1 262.0 272.0 (8.3) rescaled (to size 44) 253.6 248.3 280.0 272.9 (8.5) (11.0) (3.0) (3.9) N (after rescaling) 274 39 8 10 ◮ Pentagons/sheet rescaled using means for each size in each column. N in final row is pooled # observations for all die sizes.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Table 2: Production Costs

Input Share of Production Costs (%) Input Cost (in Rs) rexine 19.79 39.68 (5.37) (13.87) cotton/poly cloth 12.32 23.27 (4.56) (8.27) latex 13.94 38.71 (10.73) (90.71) bladder 21.07 42.02 (4.87) (14.09) labor for cutting 0.76 1.47 (0.21) (0.30) labor for stitching 19.67 39.24 (5.25) (12.82)

• ther labor (laminating, washing, packing, matching)

7.32 15.59 (4.55) (13.21)

• verhead

5.14 10.84 (2.05) (6.10) total 100.00 210.83 N 38 38 ◮ For promotional (lower-quality) ball. Data from baseline survey. Exchange rate: 100 Rs/US$1.

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Table 3: Net Benefits of Adoption

10th 25th 50th 75th 90th mean

• A. Variable cost reduction

laminated rexine waste reduction (%) 4.39 5.19 7.93 8.31 13.43 7.69 (0.46) (0.41) (0.30) (0.05) (1.18) (0.22) laminated rexine as share of cost (%) 34.85 39.87 44.72 51.22 55.44 45.94 (1.14) (0.71) (0.58) (0.44) (0.95) (0.66) variable cost reduction (%) 0.60 0.80 1.10 1.37 1.94 1.17 (0.05) (0.04) (0.03) (0.06) (0.17) (0.04)

• B. Variable cost increase

cutter wage as share of cost (%) 0.29 0.36 0.45 0.60 0.70 0.48 (0.01) (0.00) (0.01) (0.01) (0.02) (0.01) variable cost increase (%) 0.05 0.06 0.07 0.10 0.12 0.08 (0.00) (0.00) (0.00) (0.00) (0.00) (0.00)

• C. Net benefits

net variable cost reduction (%) 0.52 0.72 1.02 1.29 1.87 1.09 (0.05) (0.03) (0.03) (0.06) (0.17) (0.04) % net variable cost/avg % profit rate 5.27 8.10 12.34 19.86 28.98 15.45 (0.42) (0.50) (0.73) (1.21) (2.26) (0.71) total cost savings per month (Rs 000s) 4.46 12.19 49.38 165.21 475.01 174.12 (0.60) (1.24) (5.43) (18.42) (79.89) (18.54) days to recover fixed costs 8.48 15.98 36.61 80.34 193.92 136.94 (0.97) (1.68) (2.93) (6.51) (15.64) (44.54) days to recover fixed costs (no die) 4.40 8.30 19.01 41.71 100.69 71.10 (0.51) (0.87) (1.52) (3.38) (8.12) (23.13) ◮ Uses hot-deck imputation procedure. Standard deviation of 1,000 replications of hot-deck procedure in parentheses.

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Table 5: Treatment Assignment, Tech-Drop Experiment

# Firms Tech Drop Cash Drop No Drop Total

• A. Initial responders

smallest 5 3 12 20 medium-small 6 3 13 22 medium-large 6 3 13 22 largest 6 3 12 21 total 23 12 50 85

• B. Late responders

active, late response 12 5 14 31 active, refused all surveys 1 15 16 inactive 7 3 12 22 total 19 9 41 69

Return Baseline balance Summary statistics, initial-responder sample Summary statistics, full sample

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Table 7: Adoption as of Aug. 2013

Tech Drop Cash Drop No Drop Total

• A. Initial-responder sample

# ever active firms 23 12 50 85 # ever responded 23 12 50 85 # currently active and ever responded 22 11 46 79 # traded in 15 15 # ordered new die (beyond trade-in) 1 4 5 # received new die (beyond trade-in) 1 2 3 # ever used new die (>1000 balls) 4 4 # currently using new die (>1000 balls) 4 4

• B. Full sample

# ever active firms 35 18 79 132 # ever responded 35 17 64 116 # currently active and ever responded 32 15 59 106 # traded in 19 19 # ordered new die (beyond trade-in) 1 6 7 # received new die (beyond trade-in) 1 4 5 # ever used new die (>1000 balls) 5 1 6 # currently using new die (>1000 balls) 5 1 6

Return Characteristics of exiters Adoption as of March 2014 Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Adoption Regressions, Scale & Quality Variables

• Dep. var.: indicator for currently using offset die

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) tech drop group 0.18** 0.18** 0.59 0.17** (0.08) (0.08) (0.51) (0.07) cash drop group

• 0.00

(0.02) log avg output/month 0.03 0.04* 0.03 0.06 (0.03) (0.02) (0.03) (0.04) log avg output*tech drop

• 0.04

(0.05) share standard (of size 5)

• 0.39
• 0.38
• 0.44

(0.32) (0.33) (0.27) log avg price, size 5 training

• 0.06
• 0.19*

(0.05) (0.11) avg share promotional (of size 5)

• 0.11
• 0.16

(0.07) (0.10) avg profit rate, size 5 training 0.54 0.37 (0.64) (0.62) constant 0.02 0.02

• 0.22
• 0.29

0.41 0.14 0.40 0.11 0.02 1.14* (0.05) (0.05) (0.22) (0.18) (0.32) (0.45) (0.31) (0.07) (0.05) (0.67) stratum dummies Y Y Y Y Y Y Y Y Y Y R-squared 0.22 0.22 0.10 0.25 0.16 0.18 0.09 0.10 0.10 0.37 N 79 79 79 79 74 74 68 79 66 63

◮ Initial-responder sample. Linear probability model.

Return Full sample

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Adoption Regressions, Manager & Cutter Chars.

• Dep. var.: indicator for using offset die (> 1000 balls) as of Aug. 2013

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) tech drop group 0.18** 0.22** (0.08) (0.09) CEO university indicator 0.04 0.01 (0.07) (0.07) CEO experience (/100)

• 0.24
• 0.21

(0.17) (0.28) age of firm (/100)

• 0.06

0.03 (0.09) (0.16) cutters paid piece rate 0.02

• 0.04

(0.03) (0.05) Rs/ball, head cutter 0.11 (0.15) head cutter experience (/100)

• 0.03

(0.09) head cutter tenure (/100)

• 0.19

(0.23) cutter raven’s score

• 0.01

(0.03) avg pent/sheet, rescaled (/100) 0.65*

• 0.04

(0.37) (0.45) log avg output/month 0.05 (0.04) constant 0.02 0.05 0.11 0.07 0.04

• 0.10

0.00 0.03 0.03

• 1.58*
• 0.24

(0.05) (0.05) (0.07) (0.05) (0.04) (0.19) (0.01) (0.03) (0.07) (0.90) (0.99) stratum dummies Y Y Y Y Y Y Y Y Y Y Y R-squared 0.22 0.09 0.09 0.08 0.09 0.11 0.12 0.12 0.18 0.12 0.31 N 79 70 77 78 75 74 33 32 37 70 56

◮ Initial-responder sample. Linear probability model.

Return Full sample

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

A Model of Organizational Barriers to Adoption

◮ Goal: simplest possible model that can capture the sort of strategic communication illustrated by the anecdotes.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

A Model of Organizational Barriers to Adoption

◮ Goal: simplest possible model that can capture the sort of strategic communication illustrated by the anecdotes. ◮ Elements:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

A Model of Organizational Barriers to Adoption

◮ Goal: simplest possible model that can capture the sort of strategic communication illustrated by the anecdotes. ◮ Elements:

◮ Principal-agent model with non-contractible effort, limited

liability for agent.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

A Model of Organizational Barriers to Adoption

◮ Goal: simplest possible model that can capture the sort of strategic communication illustrated by the anecdotes. ◮ Elements:

◮ Principal-agent model with non-contractible effort, limited

liability for agent.

◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

A Model of Organizational Barriers to Adoption

◮ Goal: simplest possible model that can capture the sort of strategic communication illustrated by the anecdotes. ◮ Elements:

◮ Principal-agent model with non-contractible effort, limited

liability for agent.

◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal

◮ We show that there exist conditions under which:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

A Model of Organizational Barriers to Adoption

◮ Goal: simplest possible model that can capture the sort of strategic communication illustrated by the anecdotes. ◮ Elements:

◮ Principal-agent model with non-contractible effort, limited

liability for agent.

◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal

◮ We show that there exist conditions under which:

◮ If only standard linear contract (that cannot be conditioned on

technology type) available, a beneficial technology is not adopted.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

A Model of Organizational Barriers to Adoption

◮ Goal: simplest possible model that can capture the sort of strategic communication illustrated by the anecdotes. ◮ Elements:

◮ Principal-agent model with non-contractible effort, limited

liability for agent.

◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal

◮ We show that there exist conditions under which:

◮ If only standard linear contract (that cannot be conditioned on

technology type) available, a beneficial technology is not adopted.

◮ Truthful revelation is not an equilibrium. Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

A Model of Organizational Barriers to Adoption

◮ Goal: simplest possible model that can capture the sort of strategic communication illustrated by the anecdotes. ◮ Elements:

◮ Principal-agent model with non-contractible effort, limited

liability for agent.

◮ Strategic communication within firm

◮ Fully informed agent ◮ Imperfectly informed principal

◮ We show that there exist conditions under which:

◮ If only standard linear contract (that cannot be conditioned on

technology type) available, a beneficial technology is not adopted.

◮ Truthful revelation is not an equilibrium.

◮ If contract can be conditioned on technology type (with some

transaction cost) and transaction cost is sufficiently low, adoption occurs.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

◮ Agent faces effort cost: e(a) = a2

2 .

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

◮ Agent faces effort cost: e(a) = a2

2 .

◮ Output sells at price p.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

◮ Agent faces effort cost: e(a) = a2

2 .

◮ Output sells at price p. ◮ Materials cost: C(q) = cq

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

◮ Agent faces effort cost: e(a) = a2

2 .

◮ Output sells at price p. ◮ Materials cost: C(q) = cq ◮ Wage w(q)

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

◮ Agent faces effort cost: e(a) = a2

2 .

◮ Output sells at price p. ◮ Materials cost: C(q) = cq ◮ Wage w(q) ◮ Principal’s payoff: pq − w(q) − cq

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

◮ Agent faces effort cost: e(a) = a2

2 .

◮ Output sells at price p. ◮ Materials cost: C(q) = cq ◮ Wage w(q) ◮ Principal’s payoff: pq − w(q) − cq

◮ Wage contracts:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

◮ Agent faces effort cost: e(a) = a2

2 .

◮ Output sells at price p. ◮ Materials cost: C(q) = cq ◮ Wage w(q) ◮ Principal’s payoff: pq − w(q) − cq

◮ Wage contracts:

◮ Assumed initially to be of the form: w(q) = α + βq.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

◮ Agent faces effort cost: e(a) = a2

2 .

◮ Output sells at price p. ◮ Materials cost: C(q) = cq ◮ Wage w(q) ◮ Principal’s payoff: pq − w(q) − cq

◮ Wage contracts:

◮ Assumed initially to be of the form: w(q) = α + βq. ◮ Limited liability: α ≥ 0. Agent can’t be made to pay for job.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up

◮ Basics:

◮ Agent produces output q = sa, where s is speed of technology

and a is effort (non-contractible).

◮ Agent faces effort cost: e(a) = a2

2 .

◮ Output sells at price p. ◮ Materials cost: C(q) = cq ◮ Wage w(q) ◮ Principal’s payoff: pq − w(q) − cq

◮ Wage contracts:

◮ Assumed initially to be of the form: w(q) = α + βq. ◮ Limited liability: α ≥ 0. Agent can’t be made to pay for job. ◮ Below we will consider cases in which principal can/cannot

condition contract on ex-post-revealed characteristics of technology.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up (cont.)

◮ Technology:

◮ Existing technology has speed s0, cost per unit c0.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up (cont.)

◮ Technology:

◮ Existing technology has speed s0, cost per unit c0. ◮ New technology is one of 3 types:

• 1. c1 = c0, s1 < so: Dominated by existing technology.
• 2. c2 < c0, s2 < s0: Our technology.
• 3. c3 = c0, s3 > s0: Faster than existing technology, same cost.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up (cont.)

◮ Technology:

◮ Existing technology has speed s0, cost per unit c0. ◮ New technology is one of 3 types:

• 1. c1 = c0, s1 < so: Dominated by existing technology.
• 2. c2 < c0, s2 < s0: Our technology.
• 3. c3 = c0, s3 > s0: Faster than existing technology, same cost.

◮ Agent knows type.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up (cont.)

◮ Technology:

◮ Existing technology has speed s0, cost per unit c0. ◮ New technology is one of 3 types:

• 1. c1 = c0, s1 < so: Dominated by existing technology.
• 2. c2 < c0, s2 < s0: Our technology.
• 3. c3 = c0, s3 > s0: Faster than existing technology, same cost.

◮ Agent knows type. ◮ Principal has priors: ρ1, ρ2, ρ3, with 3

i=1 ρi = 1

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up (cont.)

◮ Technology:

◮ Existing technology has speed s0, cost per unit c0. ◮ New technology is one of 3 types:

• 1. c1 = c0, s1 < so: Dominated by existing technology.
• 2. c2 < c0, s2 < s0: Our technology.
• 3. c3 = c0, s3 > s0: Faster than existing technology, same cost.

◮ Agent knows type. ◮ Principal has priors: ρ1, ρ2, ρ3, with 3

i=1 ρi = 1

◮ Fixed cost of adoption of new technology: F

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Set-up (cont.)

◮ Technology:

◮ Existing technology has speed s0, cost per unit c0. ◮ New technology is one of 3 types:

• 1. c1 = c0, s1 < so: Dominated by existing technology.
• 2. c2 < c0, s2 < s0: Our technology.
• 3. c3 = c0, s3 > s0: Faster than existing technology, same cost.

◮ Agent knows type. ◮ Principal has priors: ρ1, ρ2, ρ3, with 3

i=1 ρi = 1

◮ Fixed cost of adoption of new technology: F

◮ Timing:

◮ Stage 1: Principal chooses wage contract. ◮ Stage 2: Agent can send one of three costless messages,

{m1, m2, m3}, about type of technology.

◮ Stage 3: Principal adopts or not. ◮ Stage 4: Payoffs realized.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Benchmark 1: Fully Informed Principal

◮ Principal’s problem: max

a,β

psia − (α + βsia) − cisia s.t. α + βsia − a2

2

≥ ¯ u (PC) arg maxa α + βsia − a2

2

= a (ICC) α ≥ 0 (LLC)

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Benchmark 1: Fully Informed Principal

◮ Principal’s problem: max

a,β

psia − (α + βsia) − cisia s.t. α + βsia − a2

2

≥ ¯ u (PC) arg maxa α + βsia − a2

2

= a (ICC) α ≥ 0 (LLC) ◮ Optimal contract: αi = 0, βi = p − ci 2

◮ Agent receives rent, because of limited liability. ◮ Bargaining model would have similar flavor.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Benchmark 2: No Signal from Agent

◮ Principal bases decision solely on priors.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Benchmark 2: No Signal from Agent

◮ Principal bases decision solely on priors. ◮ Turns out optimal contract is weighted average of optimal piece rates from full-information case: α = 0, β′ =

3

• i=1

λiβi where λi = ρis2

i

3

1 ρis2 i

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Benchmark 2: No Signal from Agent

◮ Principal bases decision solely on priors. ◮ Turns out optimal contract is weighted average of optimal piece rates from full-information case: α = 0, β′ =

3

• i=1

λiβi where λi = ρis2

i

3

1 ρis2 i

◮ Expected profit is: π′ = 3

• i=1

ρis2

i

• β′2 − F

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Parameter Restrictions

◮ Notation: πi(β) = s2

i β (p − β − ci) − F · ✶(i ∈ {1, 2, 3})

◮ β not necessarily optimal.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Parameter Restrictions

◮ Notation: πi(β) = s2

i β (p − β − ci) − F · ✶(i ∈ {1, 2, 3})

◮ β not necessarily optimal.

◮ To focus on interesting case, we impose 3 restrictions:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Parameter Restrictions

◮ Notation: πi(β) = s2

i β (p − β − ci) − F · ✶(i ∈ {1, 2, 3})

◮ β not necessarily optimal.

◮ To focus on interesting case, we impose 3 restrictions:

◮ Technology 2 more profitable than existing technology, even

under optimal piece rate for existing technology: π2(β0) > π0(β0) (1)

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Parameter Restrictions

◮ Notation: πi(β) = s2

i β (p − β − ci) − F · ✶(i ∈ {1, 2, 3})

◮ β not necessarily optimal.

◮ To focus on interesting case, we impose 3 restrictions:

◮ Technology 2 more profitable than existing technology, even

under optimal piece rate for existing technology: π2(β0) > π0(β0) (1)

◮ Technology 3 more profitable than existing technology, even

under optimal piece rate for type 2: π3(β2) > π0(β2) (2)

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Parameter Restrictions

◮ Notation: πi(β) = s2

i β (p − β − ci) − F · ✶(i ∈ {1, 2, 3})

◮ β not necessarily optimal.

◮ To focus on interesting case, we impose 3 restrictions:

◮ Technology 2 more profitable than existing technology, even

under optimal piece rate for existing technology: π2(β0) > π0(β0) (1)

◮ Technology 3 more profitable than existing technology, even

under optimal piece rate for type 2: π3(β2) > π0(β2) (2)

◮ If principal bases decision only on priors, new technology not

adopted: π0(β0) > π′ (3)

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed Principal, No Conditional Contract

◮ Proposition 1: Under (1)-(3), without conditional contracts, the following set of strategies is part of a perfect Bayesian equilibrium (PBE).

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed Principal, No Conditional Contract

◮ Proposition 1: Under (1)-(3), without conditional contracts, the following set of strategies is part of a perfect Bayesian equilibrium (PBE).

◮ Agent’s strategy:

◮ If the technology is of type 1 or type 2, signal m1 ◮ If the technology is of type 3, signal m3. Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed Principal, No Conditional Contract

◮ Proposition 1: Under (1)-(3), without conditional contracts, the following set of strategies is part of a perfect Bayesian equilibrium (PBE).

◮ Agent’s strategy:

◮ If the technology is of type 1 or type 2, signal m1 ◮ If the technology is of type 3, signal m3.

◮ Principal’s strategy:

◮ Offer wage contract

• α∗ = 0, β∗ = p−c0

2

• ◮ If agent signals m2 or m3, adopt.

◮ If agent signals m1, do not adopt. Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed Principal, No Conditional Contract

◮ Proposition 1: Under (1)-(3), without conditional contracts, the following set of strategies is part of a perfect Bayesian equilibrium (PBE).

◮ Agent’s strategy:

◮ If the technology is of type 1 or type 2, signal m1 ◮ If the technology is of type 3, signal m3.

◮ Principal’s strategy:

◮ Offer wage contract

• α∗ = 0, β∗ = p−c0

2

• ◮ If agent signals m2 or m3, adopt.

◮ If agent signals m1, do not adopt.

◮ Intuition:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed Principal, No Conditional Contract

◮ Proposition 1: Under (1)-(3), without conditional contracts, the following set of strategies is part of a perfect Bayesian equilibrium (PBE).

◮ Agent’s strategy:

◮ If the technology is of type 1 or type 2, signal m1 ◮ If the technology is of type 3, signal m3.

◮ Principal’s strategy:

◮ Offer wage contract

• α∗ = 0, β∗ = p−c0

2

• ◮ If agent signals m2 or m3, adopt.

◮ If agent signals m1, do not adopt.

◮ Intuition:

◮ Given a fixed-ex-ante piece rate, agent clearly prefers faster

technology, regardless of fixed cost.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed Principal, No Conditional Contract

◮ Proposition 1: Under (1)-(3), without conditional contracts, the following set of strategies is part of a perfect Bayesian equilibrium (PBE).

◮ Agent’s strategy:

◮ If the technology is of type 1 or type 2, signal m1 ◮ If the technology is of type 3, signal m3.

◮ Principal’s strategy:

◮ Offer wage contract

• α∗ = 0, β∗ = p−c0

2

• ◮ If agent signals m2 or m3, adopt.

◮ If agent signals m1, do not adopt.

◮ Intuition:

◮ Given a fixed-ex-ante piece rate, agent clearly prefers faster

technology, regardless of fixed cost.

◮ But why is principal influenced by the agent’s signal, given

that she knows agent has this incentive?

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed Principal, No Conditional Contract

◮ Proposition 1: Under (1)-(3), without conditional contracts, the following set of strategies is part of a perfect Bayesian equilibrium (PBE).

◮ Agent’s strategy:

◮ If the technology is of type 1 or type 2, signal m1 ◮ If the technology is of type 3, signal m3.

◮ Principal’s strategy:

◮ Offer wage contract

• α∗ = 0, β∗ = p−c0

2

• ◮ If agent signals m2 or m3, adopt.

◮ If agent signals m1, do not adopt.

◮ Intuition:

◮ Given a fixed-ex-ante piece rate, agent clearly prefers faster

technology, regardless of fixed cost.

◮ But why is principal influenced by the agent’s signal, given

that she knows agent has this incentive?

◮ Agent’s and principal’s interests are aligned for types 1 and 3.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed Principal, No Conditional Contract

◮ Proposition 1: Under (1)-(3), without conditional contracts, the following set of strategies is part of a perfect Bayesian equilibrium (PBE).

◮ Agent’s strategy:

◮ If the technology is of type 1 or type 2, signal m1 ◮ If the technology is of type 3, signal m3.

◮ Principal’s strategy:

◮ Offer wage contract

• α∗ = 0, β∗ = p−c0

2

• ◮ If agent signals m2 or m3, adopt.

◮ If agent signals m1, do not adopt.

◮ Intuition:

◮ Given a fixed-ex-ante piece rate, agent clearly prefers faster

technology, regardless of fixed cost.

◮ But why is principal influenced by the agent’s signal, given

that she knows agent has this incentive?

◮ Agent’s and principal’s interests are aligned for types 1 and 3. ◮ Signal about type 3 is valuable to principal.

Sketch of proof Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed, No Conditional Contracts (cont.)

◮ There are other equilibria, including “babbling” equilibrium: principal ignores signal from agent, agent babbles.

◮ Under (3), principal does not adopt.

Sketch of proof Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed, No Conditional Contracts (cont.)

◮ There are other equilibria, including “babbling” equilibrium: principal ignores signal from agent, agent babbles.

◮ Under (3), principal does not adopt.

◮ Proposition 2: Under (1)-(3), without conditional contracts, there is no perfect Bayesian equilibrium under which the agent always truthfully reveals the technology type.

Sketch of proof Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed, No Conditional Contracts (cont.)

◮ There are other equilibria, including “babbling” equilibrium: principal ignores signal from agent, agent babbles.

◮ Under (3), principal does not adopt.

◮ Proposition 2: Under (1)-(3), without conditional contracts, there is no perfect Bayesian equilibrium under which the agent always truthfully reveals the technology type. ◮ Intuition:

Sketch of proof Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed, No Conditional Contracts (cont.)

◮ There are other equilibria, including “babbling” equilibrium: principal ignores signal from agent, agent babbles.

◮ Under (3), principal does not adopt.

◮ Proposition 2: Under (1)-(3), without conditional contracts, there is no perfect Bayesian equilibrium under which the agent always truthfully reveals the technology type. ◮ Intuition:

◮ Suppose that in Stage 1, the principal chooses optimal piece

rate expecting the agent to reveal truthfully.

Sketch of proof Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed, No Conditional Contracts (cont.)

◮ There are other equilibria, including “babbling” equilibrium: principal ignores signal from agent, agent babbles.

◮ Under (3), principal does not adopt.

◮ Proposition 2: Under (1)-(3), without conditional contracts, there is no perfect Bayesian equilibrium under which the agent always truthfully reveals the technology type. ◮ Intuition:

◮ Suppose that in Stage 1, the principal chooses optimal piece

rate expecting the agent to reveal truthfully.

◮ If technology is of type 2, given fixed piece rate, agent will

want to deviate and report that it is type 1, dissuading principal from adopting.

Sketch of proof Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed, Conditional Contracts

◮ Now assume principal can pay a fixed cost, G, to be able to

• ffer contracts of form:

w(q) = α + βq + γq if c = c2 w(q) = α + βq if c = c2

◮ Note that type 2 is only technology for which the optimal

piece rate differs from that for existing technology.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Imperfectly Informed, Conditional Contracts (cont.)

Proposition 3: Under (1)-(3), with conditional contracts, if G < ρ2 [π2(β2) − π0(β0)] (4) then the following set of strategies is part of a PBE: ◮ Agent’s strategy:

◮ If the principal pays G, signal truthfully. ◮ If the principal does not pay G:

◮ If the technology is of type 1 or 2, signal m1. ◮ If the technology is of type 3, signal m3.

◮ Principal’s strategy:

◮ Pay G and offer wage contract

• α∗∗ = 0, β∗∗ = p − c0

2 , γ∗∗ = c0 − c2 2

• ◮ If the agent signals m2 or m3, adopt.

◮ If the agent signals m1, do not adopt.

Sketch of proof Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Model: Discussion

◮ Why doesn’t principal just adopt conditional contract? Model suggests two reasons:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Model: Discussion

◮ Why doesn’t principal just adopt conditional contract? Model suggests two reasons:

◮ Principal just not aware of alternative contract.

◮ World of Proposition 1. Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Model: Discussion

◮ Why doesn’t principal just adopt conditional contract? Model suggests two reasons:

◮ Principal just not aware of alternative contract.

◮ World of Proposition 1.

◮ Principal is aware of alternative contract, expected benefit less

than cost of offering it.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Model: Discussion

◮ Why doesn’t principal just adopt conditional contract? Model suggests two reasons:

◮ Principal just not aware of alternative contract.

◮ World of Proposition 1.

◮ Principal is aware of alternative contract, expected benefit less

than cost of offering it.

◮ If condition on fixed cost in Proposition 3

G < ρ2 [π2(β2) − π0(β0)] not satisfied, then misinformation, non-adoption of type 2 is still an equilibrium.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Model: Discussion

◮ Why doesn’t principal just adopt conditional contract? Model suggests two reasons:

◮ Principal just not aware of alternative contract.

◮ World of Proposition 1.

◮ Principal is aware of alternative contract, expected benefit less

than cost of offering it.

◮ If condition on fixed cost in Proposition 3

G < ρ2 [π2(β2) − π0(β0)] not satisfied, then misinformation, non-adoption of type 2 is still an equilibrium.

◮ Will not be satisfied if ρ2 sufficiently low. Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Model: Discussion

◮ Why doesn’t principal just adopt conditional contract? Model suggests two reasons:

◮ Principal just not aware of alternative contract.

◮ World of Proposition 1.

◮ Principal is aware of alternative contract, expected benefit less

than cost of offering it.

◮ If condition on fixed cost in Proposition 3

G < ρ2 [π2(β2) − π0(β0)] not satisfied, then misinformation, non-adoption of type 2 is still an equilibrium.

◮ Will not be satisfied if ρ2 sufficiently low.

◮ Model is one owner/one cutter, but heterogeneity in adoption could be explained by heterogeneity in ρ2.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 1

◮ Show that neither agent nor principal has incentive to deviate. ◮ Agent’s decision, holding principal’s strategy fixed:

◮ If type 1, no incentive to deviate from m1, since m2 or m3

would induce adoption and s1 < s0.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 1

◮ Show that neither agent nor principal has incentive to deviate. ◮ Agent’s decision, holding principal’s strategy fixed:

◮ If type 1, no incentive to deviate from m1, since m2 or m3

would induce adoption and s1 < s0.

◮ If type 2, no incentive to deviate from m1, since m2 or m3

would induce adoption and s2 < s0.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 1

◮ Show that neither agent nor principal has incentive to deviate. ◮ Agent’s decision, holding principal’s strategy fixed:

◮ If type 1, no incentive to deviate from m1, since m2 or m3

would induce adoption and s1 < s0.

◮ If type 2, no incentive to deviate from m1, since m2 or m3

would induce adoption and s2 < s0.

◮ If type 3, no incentive to deviate from m3, since (a) m1 would

discourage adoption and s0 < s3; (b) no gain from m2.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 1

◮ Show that neither agent nor principal has incentive to deviate. ◮ Agent’s decision, holding principal’s strategy fixed:

◮ If type 1, no incentive to deviate from m1, since m2 or m3

would induce adoption and s1 < s0.

◮ If type 2, no incentive to deviate from m1, since m2 or m3

would induce adoption and s2 < s0.

◮ If type 3, no incentive to deviate from m3, since (a) m1 would

discourage adoption and s0 < s3; (b) no gain from m2.

◮ Principal’s decision, holding agent’s strategy fixed:

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 1

◮ Show that neither agent nor principal has incentive to deviate. ◮ Agent’s decision, holding principal’s strategy fixed:

◮ If type 1, no incentive to deviate from m1, since m2 or m3

would induce adoption and s1 < s0.

◮ If type 2, no incentive to deviate from m1, since m2 or m3

would induce adoption and s2 < s0.

◮ If type 3, no incentive to deviate from m3, since (a) m1 would

discourage adoption and s0 < s3; (b) no gain from m2.

◮ Principal’s decision, holding agent’s strategy fixed:

◮ If receive m2 or m3, no incentive to deviate and not adopt,

since: (ρ1 + ρ2)π0(β∗) + ρ3π3(β∗) > π0(β0) since s3 > s0, c3 = c0.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 1

◮ Show that neither agent nor principal has incentive to deviate. ◮ Agent’s decision, holding principal’s strategy fixed:

◮ If type 1, no incentive to deviate from m1, since m2 or m3

would induce adoption and s1 < s0.

◮ If type 2, no incentive to deviate from m1, since m2 or m3

would induce adoption and s2 < s0.

◮ If type 3, no incentive to deviate from m3, since (a) m1 would

discourage adoption and s0 < s3; (b) no gain from m2.

◮ Principal’s decision, holding agent’s strategy fixed:

◮ If receive m2 or m3, no incentive to deviate and not adopt,

since: (ρ1 + ρ2)π0(β∗) + ρ3π3(β∗) > π0(β0) since s3 > s0, c3 = c0.

◮ If receive m1, no incentive to deviate and adopt. In this case,

would set β based solely on priors in Stage 1. Under condition (3), better off not adopting.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 2

◮ Proof by contradiction:

◮ Assume that there is a PBE with truthful revelation. ◮ Under (1)-(3), principal will adopt if agent signals m2 or m3. ◮ But when technology is of type 2, for any choice of piece rate

β by principal, agent can do better by deviating and signalling m1: U(β, s0) = (β∗s0)2 2 > (β∗s2)2 2 = U(β, s2) ∀ β which contradicts assumption that truthful revelation is part of a PBE.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 3

◮ Show that neither agent nor principal has incentive to deviate. ◮ Agent’s decision, holding principal’s strategy fixed:

◮ If type 1, no incentive to deviate from m1, since m2 or m3

would induce adoption but not change piece rate. Since s1 < s0, agent would be worse off.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 3

◮ Show that neither agent nor principal has incentive to deviate. ◮ Agent’s decision, holding principal’s strategy fixed:

◮ If type 1, no incentive to deviate from m1, since m2 or m3

would induce adoption but not change piece rate. Since s1 < s0, agent would be worse off.

◮ If type 2, no incentive to deviate from m2 to m1 if:

((β∗∗ + γ∗∗)s2)2 2 > (β∗∗s0)2 2 Condition 1 implies this. No gain from deviating to m3.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 3

◮ Show that neither agent nor principal has incentive to deviate. ◮ Agent’s decision, holding principal’s strategy fixed:

◮ If type 1, no incentive to deviate from m1, since m2 or m3

would induce adoption but not change piece rate. Since s1 < s0, agent would be worse off.

◮ If type 2, no incentive to deviate from m2 to m1 if:

((β∗∗ + γ∗∗)s2)2 2 > (β∗∗s0)2 2 Condition 1 implies this. No gain from deviating to m3.

◮ If type 3, no incentive to deviate from m3, since (a) m1 would

discourage adoption and s0 < s3; (b) no gain from m2.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 3 (cont.)

◮ Principal’s decision, holding agent’s strategy fixed:

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 3 (cont.)

◮ Principal’s decision, holding agent’s strategy fixed:

◮ Under (1)-(3), if G is paid then strategy is optimal.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Sketch of Proof of Proposition 3 (cont.)

◮ Principal’s decision, holding agent’s strategy fixed:

◮ Under (1)-(3), if G is paid then strategy is optimal. ◮ If G is not paid, then optimal strategy for principal is the one

in Proposition 1. There is no incentive to deviate to this strategy if: ρ1π0(β0)+ρ2π2(β2)+ρ3π3(β3)−G > (ρ1 + ρ2) π0(β∗)+ρ3π3(β∗) which holds if and only if (4) holds, since β0 = β3.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Incentive Payment Experiment: “Test” Results

firm 1 2 3 4 5 6 7 8 9 10 time 2:52 2:40 3:03 3:02 2:59 2:28 2:25 2:45 2:30 2:50 die size 43.5 43.75 44 44 43.5 43.5 43.5 43.5 44 43.5 # pentagons 270 272 273 272 282 279 279 272 272 267

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

• Fig. 16: Permutation Test, Current Use as Outcome

.1 .2 .3 .4 Fraction

• .4
• .2

.2 .4 Coefficient from a Permutation Outcome

Histogram of 25,871,920 possible permutation outcomes.

All Strata

.1 .2 .3 .4 Fraction

• .4
• .2

.2 .4 Coefficient from a Permutation Outcome

Histogram of 1,293,600 possible permutation outcomes.

Initial Non-Adopters

Vertical line denotes the observed regression coefficient.

◮ Notes: Figure displays distribution of ITT regression coefficients from all possible assignments to treatment. Left panel includes all firms. Right panel includes only initial non-adopters.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

• Fig. 17: Permutation Test, Die Purchase as Outcome

.05 .1 .15 .2 Fraction −.4 −.2 .2 .4 Coefficient from a Permutation Outcome

Histogram of 25,871,920 possible permutation outcomes.

All Strata

.1 .2 .3 .4 Fraction −.2 −.1 .1 .2 Coefficient from a Permutation Outcome

Histogram of 1,293,600 possible permutation outcomes.

Initial Non−Adopters

Vertical line denotes the observed regression coefficient.

◮ Notes: Figure displays distribution of ITT regression coefficients from all possible assignments to treatment. Left panel includes all firms. Right panel includes only initial non-adopters.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Likelihood Function

l(θ, σε) =

• f
• (1 − adopt1f )(adopt2f ) ln
• Φ

ln(NVBf + 320) − θ σε

• −Φ

ln(NVBf ) − θ σε

• +(1 − adopt1f )(1 − adopt2f ) ln
• 1 − Φ

ln(NVBf + 320) − θ σε

• +(adopt1f ) ln
• Φ

ln(NVBf ) − θ σε

• where Φ(·)=cdf of standard normal.

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Competitive Pressures

◮ Squeezed on high end:

◮ Adidas Jabulani thermo-molding technology, introduced for

2006 World Cup in Thailand, produced in China for 2010 World Cup.

◮ One firm in Sialkot (Forward Sports) now has technology,

produced thermo-molded balls for 2012 Euro Cup.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Competitive Pressures

◮ Squeezed on high end:

◮ Adidas Jabulani thermo-molding technology, introduced for

2006 World Cup in Thailand, produced in China for 2010 World Cup.

◮ One firm in Sialkot (Forward Sports) now has technology,

produced thermo-molded balls for 2012 Euro Cup.

◮ Squeezed on low end:

◮ Machine-stitching technology, also mainly made in China.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Competitive Pressures

◮ Squeezed on high end:

◮ Adidas Jabulani thermo-molding technology, introduced for

2006 World Cup in Thailand, produced in China for 2010 World Cup.

◮ One firm in Sialkot (Forward Sports) now has technology,

produced thermo-molded balls for 2012 Euro Cup.

◮ Squeezed on low end:

◮ Machine-stitching technology, also mainly made in China.

◮ Still primary location for production of top-quality hand-stitched balls.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

U.S. Imports of Soccer Balls (Footballs), 1989-2013

10 20 30 40 50 total imports (fob) into US (mil. current US$ 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 year

Pakistan China Indonesia Vietnam India Thailand Korea Taiwan Other

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Production Costs (pilot firms)

Firm 1 Costs (Rs) Firm 2 Costs (Rs) Rexine 50 42.3 Cotton/polyester cloth 30 40 Latex 75 74 Labor for lamination 1.55 2 Labor for cutting 1 2 Labor for printing, matching 8 15 Labor for stitching 45 40 Bladder 52 60 Other (checking, washing, packing, overhead) 17 20 Total 279.55 295.30 ◮ 100 Rs ≈ US$1. ◮ Both firms mid-sized, middle of quality range. ◮ Laminated rexine cost (rexine + cloth + latex) is 53-55% of total. Labor for cutting .4-.7%.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Timeline

◮ Census (listing, no survey): Sept.-Nov. 2011 ◮ Baseline: Jan.-April 2012 ◮ Round 0: May-June 2012

◮ Technology drop ◮ Cash drop ◮ Follow-up with non-treated firms.

◮ Round 1: July 2012

◮ Follow-ups with all firms.

◮ Round 2: Oct. 2012 ◮ Round 3: Jan. 2013 ◮ Round 4: March-April 2013 ◮ Round 5, incentive-payment experiment: Sept.-Nov. 2013 ◮ Round 6, Jan.-March 2014

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Quantiles, Initial-Responder Sample

Mean Min 10th 25th 50th 75th 90th Max N avg output/month (000s) 32.2 0.8 1.6 3.5 10.0 34.6 83.0 275.0 85 avg employment 90.2 3.3 5.2 7.4 20.0 52.9 235.0 1,700.0 85 avg employment (cutters) 5.8 0.5 1.0 1.0 2.2 5.0 13.0 123.0 85 avg Rs/ball (head cutter) 1.5 1.0 1.1 1.3 1.5 1.6 1.9 2.9 79 avg % promotional (of size 5) 41.4 0.0 2.0 18.8 41.1 62.4 80.0 100.0 85 avg price, size 5 promotional 241.3 152.5 185.0 196.3 227.1 266.8 300.0 575.0 64 avg price, size 5 training 440.0 200.0 275.0 313.8 381.3 488.0 600.0 2,250.0 72 avg profit %, size 5 promo 8.2 2.5 3.9 5.2 8.1 10.2 12.5 20.0 64 avg profit %, size 5 training 8.0 1.6 3.2 4.6 8.5 9.9 12.5 22.2 70 avg % lamination in-house 95.7 31.3 81.3 100.0 100.0 100.0 100.0 100.0 75 % standard design (of size 5) 90.8 0.0 70.0 85.0 100.0 100.0 100.0 100.0 81 age of firm 25.4 2.0 6.0 12.0 19.5 36.5 54.0 108.0 84 CEO experience 17.0 3.0 6.0 9.0 15.5 22.0 28.0 66.0 82 head cutter experience 20.5 2.0 8.0 12.0 18.5 26.5 41.0 46.0 36 head cutter tenure 11.1 0.0 2.0 6.0 9.0 15.0 22.0 46.0 35

◮ Variables marked “avg” are averaged within firms across rounds before calculating

quantiles.

Full sample Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Quantiles, All-Responder Sample

Mean Min 10th 25th 50th 75th 90th Max N avg output/month (000s) 34.6 0.0 2.0 4.5 15.0 37.2 86.3 278.6 116 avg employment 103.9 3.3 5.6 8.0 25.0 75.0 230.0 2,180.0 115 avg employment (cutters) 5.4 0.5 1.0 1.2 2.8 5.0 12.4 123.0 114 avg Rs/ball (head cutter) 1.5 1.0 1.0 1.3 1.5 1.6 2.0 3.0 107 avg % promotional (of size 5) 37.0 0.0 0.0 8.3 33.8 55.2 80.0 100.0 114 avg price, size 5 promotional 245.7 150.0 185.0 202.0 235.0 270.0 300.0 575.0 81 avg price, size 5 training 465.0 200.0 286.7 330.0 400.0 506.8 667.9 2,250.0 100 avg profit (%), size 5 promo 8.3 2.5 4.1 5.1 7.7 10.4 13.8 20.0 80 avg profit (%), size 5 training 8.3 1.6 3.4 5.1 8.5 10.0 13.0 22.2 95 avg % lamination in-house 96.2 25.0 85.0 100.0 100.0 100.0 100.0 100.0 104

◮ Variables marked “avg” are averaged within firms across rounds before calculating

quantiles.

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Means by Firm Size Bins, Initial-Responder Sample

Firm Size Bins 1 2 3 4 avg output/month (000s) 5.43 6.18 24.49 93.08 avg employment 11.68 13.29 53.07 284.43 avg employment (cutters) 1.25 1.79 3.84 16.36 cutters paid piece rate indicator 0.90 1.00 0.91 0.79 avg Rs/ball (head cutter) 1.53 1.54 1.51 1.38 avg % promotional (of size 5) 49.44 51.40 34.47 30.61 avg price, size 5 promotional 239.57 223.76 249.23 254.26 avg price, size 5 training 387.09 329.23 442.18 617.36 avg profit %, size 5 promo 6.15 7.20 9.58 10.16 avg profit %, size 5 training 6.95 7.00 8.25 9.86 avg % lamination in-house 90.64 92.74 99.77 99.82 % standard design (of size 5) 89.00 94.43 90.43 89.21 age of firm 16.95 20.09 24.67 39.81 CEO experience 19.00 16.55 15.75 16.85 head cutter experience 13.83 20.44 26.82 17.60 head cutter tenure 12.50 7.33 13.55 11.00 N 20 22 22 21

◮ Bins defined by quartiles of output in normal month from baseline survey. Correspond to strata in randomization.

Full sample

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Means by Firm Size Bins, All-Responder Sample

Firm Size Bins Late 1 2 3 4 Responders avg output/month (000s) 5.43 6.18 24.49 93.08 41.23 avg employment 11.68 13.29 53.07 284.43 142.65 avg employment (cutters) 1.25 1.79 3.84 16.36 4.42 avg Rs/ball (head cutter) 1.53 1.54 1.51 1.38 1.61 avg % promotional (of size 5) 49.44 51.40 34.47 30.61 23.93 avg price, size 5 promotional 239.57 223.76 249.23 254.26 262.34 avg price, size 5 training 387.09 329.23 442.18 617.36 529.49 avg profit %, size 5 promo 6.15 7.20 9.58 10.16 8.68 avg profit %, size 5 training 6.95 7.00 8.25 9.86 9.29 avg % lamination in-house 90.64 92.74 99.77 99.82 97.41 N 20 22 22 21 31

◮ Bins defined by quartiles of output in normal month from baseline survey. ◮ Bins correspond to strata in randomization.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Baseline Balance, Initial-Responder Sample

Tech Drop Cash Drop No Drop Output, normal month (000s) 34.18 26.69 41.56 (11.48) (12.15) (9.53) Output, previous year (000s) 680.17 579.97 763.33 (220.13) (225.13) (232.95) Employment, normal month 42.26 82.58 92.62 (13.25) (47.16) (35.77) % size 5 84.61 88.96 82.67 (5.38) (4.52) (3.74) % promotional (of size 5) 50.12 66.09 59.02 (7.12) (11.04) (5.17) Age of firm 22.70 29.25 25.76 (2.25) (4.88) (3.09) CEO experience 16.22 20.42 16.55 (2.39) (2.70) (1.62) CEO college indicator 0.43 0.27 0.40 (0.11) (0.14) (0.08) Head cutter experience 17.00 30.33 20.91 (2.08) (6.69) (2.68) Head cutter tenure 12.20 12.00 10.50 (2.21) (5.77) (2.11) Share cutters paid piece rate 0.95 0.83 0.89 (0.05) (0.11) (0.05) Rupees/ball (head cutter) 1.44 1.63 1.37 (0.14) (0.21) (0.10) N 23 12 50

◮ No differences significant at 5% level.

Late responders Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Baseline Balance, Late Responders

Group 4 Tech Drop Group 5 Cash Drop Group 6 No Drop Output, normal month (000s) 27.85 34.80 63.13 (14.01) (4.99) (18.25) Employment, normal month 67.20 61.00 353.38 (48.18) (34.94) (264.52) % size 5 68.00 72.22 96.88 (9.80) (16.16) (3.13) % promotional (of size 5) 31.17 36.11 24.22 (9.77) (12.58) (13.28) Age of firm 17.40 39.60 35.13 (3.13) (16.68) (5.55) N 10 5 8

◮ Table uses only information from abridged baseline. ◮ 8 late responders never provided baseline information.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Adoption as of March 2014

Tech Drop Cash Drop No Drop Total

• A. Initial-responder sample

# ever active firms 23 12 50 85 # ever responded 23 12 50 85 # currently active and ever responded 22 11 46 79 # traded in 15 15 # ordered new die (beyond trade-in) 3 5 8 # received new die (beyond trade-in) 3 4 7 # ever used new die (>1000 balls) 8 1 9 # currently using new die (>1000 balls) 8 1 9

• B. All-responder sample

# ever active firms 35 18 79 132 # ever responded 35 17 64 116 # currently active and ever responded 31 15 59 105 # traded in 19 19 # ordered new die (beyond trade-in) 6 7 13 # received new die (beyond trade-in) 5 6 11 # ever used new die (>1000 balls) 10 2 12 # currently using new die (>1000 balls) 10 2 12

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Correlates of Exit, Initial-Responder Sample, Aug. 2013

Cash/No Drop Tech Drop Non-Exiter Exiter Non-Exiter Exiter log avg output/month 9.35 8.27 9.56 8.78 (0.20) (0.58) (0.31) log avg employment 3.29 2.46 3.20 2.37 (0.20) (0.61) (0.26) % standard (of size 5) 91.69 88.40 89.45 80.00 (1.76) (7.91) (4.78) CEO university indicator 0.38 0.33 0.45 0.00 (0.07) (0.33) (0.11) CEO experience 17.53 14.75 16.05 20.00 (1.51) (1.89) (2.50) age of firm 26.95 20.80 23.41 7.00 (2.84) (5.93) (2.23) log avg price, size 5 promo 5.49 5.29 5.41 5.29 (0.04) (0.04) (0.05) log avg price, size 5 training 6.06 5.62 5.93 5.99 (0.06) (0.16) (0.05) avg % promotional (of size 5) 41.05 54.20 39.48 41.11 (3.74) (20.08) (5.66) avg profit %, size 5 training 0.08 0.03 0.09 0.02 (0.01) (0.00) (0.01) head cutter experience 22.59 18.00 17.00 (2.74) (7.00) (2.08) head cutter tenure 11.27 6.33 12.20 (2.18) (0.33) (2.21) avg Rs/ball, head cutter 1.47 1.17 1.60 1.33 (0.04) (0.12) (0.09) cutter raven’s score 1.96 1.33 1.92 (0.19) (0.33) (0.19) N 57 5 22 1 Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Correlates of Exit, All-Responder Sample, Aug. 2013

Cash/No Drop Tech Drop Non-Exiter Exiter Non-Exiter Exiter log avg output/month 9.61 8.66 9.57 9.31 (0.17) (0.49) (0.24) (0.52) log avg employment 3.54 2.46 3.26 2.63 (0.18) (0.44) (0.22) (0.26) log avg price, size 5 promo 5.51 5.33 5.42 5.29 (0.04) (0.05) (0.04) log avg price, size 5 training 6.12 5.81 5.96 6.29 (0.05) (0.15) (0.04) (0.29) avg % promotional (of size 5) 38.20 38.71 34.80 20.56 (3.46) (17.08) (4.78) (20.56) avg profit %, size 5 train 0.08 0.06 0.09 0.02 (0.01) (0.02) (0.01) (0.01) avg Rs/ball, head cutter 1.49 1.20 1.58 2.28 (0.04) (0.10) (0.08) (0.49) N 74 7 32 2 Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Correlates of Exit, Initial-Responder Sample, March 2013

Cash/No Drop Tech Drop Non-Exiter Exiter Non-Exiter Exiter log avg output/month 9.35 8.27 9.56 8.78 (0.20) (0.58) (0.31) log avg employment 3.29 2.46 3.20 2.37 (0.20) (0.61) (0.26) % standard (of size 5) 91.83 88.40 89.45 80.00 (1.73) (7.91) (4.78) CEO university indicator 0.38 0.33 0.45 0.00 (0.07) (0.33) (0.11) CEO experience 17.53 14.75 16.05 20.00 (1.51) (1.89) (2.50) age of firm 26.95 20.80 23.41 7.00 (2.84) (5.93) (2.23) log avg price, size 5 promo 5.49 5.29 5.41 5.29 (0.04) (0.04) (0.05) log avg price, size 5 training 6.06 5.62 5.93 5.99 (0.06) (0.16) (0.05) avg % promotional (of size 5) 41.05 54.20 39.48 41.11 (3.74) (20.08) (5.66) avg profit %, size 5 training 0.08 0.03 0.09 0.02 (0.01) (0.00) (0.01) head cutter experience 22.59 18.00 17.00 (2.74) (7.00) (2.08) head cutter tenure 11.27 6.33 12.20 (2.18) (0.33) (2.21) avg Rs/ball, head cutter 1.47 1.17 1.60 1.33 (0.04) (0.12) (0.09) cutter raven’s score 1.96 1.33 1.92 (0.19) (0.33) (0.19) N 57 5 22 1 Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Correlates of Exit, All-Responder Sample, March 2013

Cash/No Drop Tech Drop Non-Exiter Exiter Non-Exiter Exiter log avg output/month 9.61 8.66 9.62 8.88 (0.17) (0.49) (0.25) (0.52) log avg employment 3.54 2.46 3.29 2.59 (0.18) (0.44) (0.22) (0.16) log avg price, size 5 promo 5.51 5.33 5.43 5.30 (0.04) (0.05) (0.04) (0.01) log avg price, size 5 training 6.12 5.81 5.96 6.13 (0.05) (0.15) (0.04) (0.23) avg % promotional (of size 5) 38.20 38.71 33.43 39.50 (3.46) (17.08) (4.73) (22.35) avg profit %, size 5 train 0.08 0.06 0.09 0.04 (0.01) (0.02) (0.01) (0.02) avg Rs/ball, head cutter 1.49 1.20 1.57 2.13 (0.04) (0.10) (0.08) (0.38) N 74 7 31 3 Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Exit Probit, Initial-Responder Sample

• Dep. var.: indicator for exit (as of Aug. 2013)

(1) (2) (3) (4) (5) (6) (7) (8) (9) tech drop group

• 0.31

0.69 (0.52) (0.67) log avg output/month

• 0.27*
• 0.42

(0.14) (0.55) CEO university indicator

• 0.32

(0.53) age of firm

• 0.01

0.01 (0.01) (0.01) log avg price, size 5 promo

• 0.01***
• 0.00

(0.00) (0.01) log avg price, size 5 training

• 0.01*

(0.00) avg % promotional (of size 5) 0.01

• 0.00

(0.01) (0.02) avg profit %, size 5 train

• 0.93***
• 0.85**

(0.32) (0.34) Pseudo R-squared 0.01 0.07 0.01 0.03 0.12 0.12 0.02 0.53 0.55 N 85 85 74 84 64 72 85 70 63

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Exit Probit, All-Responder Sample

• Dep. var.: indicator for exit (as of Aug. 2013)

(1) (2) (3) (4) (5) (6) (7) (8) (9) tech drop group

• 0.00
• 0.06

(0.36) (0.63) log avg output/month

• 0.21*
• 0.46*

(0.11) (0.25) CEO university indicator

• 0.33

(0.53) age of firm

• 0.03*
• 0.01

(0.01) (0.02) log avg price, size 5 promo

• 0.01**
• 0.01

(0.00) (0.01) log avg price, size 5 training

• 0.00

(0.00) avg % promotional (of size 5)

• 0.00
• 0.02

(0.01) (0.02) avg profit %, size 5 train

• 0.13
• 0.25**

(0.09) (0.11) Pseudo R-squared 0.00 0.04 0.01 0.07 0.09 0.01 0.00 0.11 0.35 N 116 115 75 107 81 100 114 95 77

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Exit Probit, Initial-Responder Sample

• Dep. var.: indicator for exit (as of March 2014)

(1) (2) (3) (4) (5) (6) (7) (8) (9) tech drop group

• 0.31

0.69 (0.52) (0.67) log avg output/month

• 0.27*
• 0.42

(0.14) (0.55) CEO university indicator

• 0.32

(0.53) age of firm

• 0.01

0.01 (0.01) (0.01) log avg price, size 5 promo

• 0.01***
• 0.00

(0.00) (0.01) log avg price, size 5 training

• 0.01*

(0.00) avg % promotional (of size 5) 0.01

• 0.00

(0.01) (0.02) avg profit %, size 5 train

• 0.93***
• 0.85**

(0.32) (0.34) Pseudo R-squared 0.01 0.07 0.01 0.03 0.12 0.12 0.02 0.53 0.55 N 85 85 74 84 64 72 85 70 63

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Exit Probit, All-Responder Sample

• Dep. var.: indicator for exit (as of March 2014)

(1) (2) (3) (4) (5) (6) (7) (8) (9) tech drop group 0.16 0.56 (0.34) (0.71) log avg output/month

• 0.24**
• 0.54

(0.11) (0.34) CEO university indicator

• 0.00

(0.46) age of firm

• 0.03*
• 0.04

(0.02) (0.03) log avg price, size 5 promo

• 0.01***
• 0.01

(0.00) (0.01) log avg price, size 5 training

• 0.00

(0.00) avg % promotional (of size 5) 0.00

• 0.01

(0.01) (0.02) avg profit %, size 5 train

• 0.13
• 0.20**

(0.08) (0.10) Pseudo R-squared 0.00 0.06

• 0.00

0.09 0.10 0.02 0.00 0.10 0.33 N 116 115 75 107 81 100 114 95 77

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Correlates of Adoption, Initial-Responder Sample, Aug. 2013

Cash/No Drop Tech Drop Non-Adopter Adopter Non-Adopter Adopter log avg output/month 9.35 9.69 8.97 (0.20) (0.36) (0.46) log avg employment 3.29 3.34 2.61 (0.20) (0.31) (0.27) % standard (of size 5) 91.83 92.67 75.00 (1.73) (2.59) (25.00) CEO university indicator 0.38 0.50 0.25 (0.07) (0.13) (0.25) CEO experience 17.53 17.00 11.75 (1.51) (2.94) (3.38) age of firm 26.95 24.89 16.75 (2.84) (2.47) (4.15) log avg price, size 5 promo 5.49 5.43 5.34 (0.04) (0.06) (0.09) log avg price, size 5 training 6.06 5.98 5.73 (0.06) (0.06) (0.12) avg % promotional (of size 5) 41.05 40.62 34.39 (3.74) (6.74) (8.16) avg profit %, size 5 training 0.08 0.09 0.09 (0.01) (0.01) (0.01) head cutter experience 22.59 16.80 19.00 (2.74) (2.29) head cutter tenure 11.27 13.22 3.00 (2.18) (2.18) avg Rs/ball, head cutter 1.47 1.55 1.80 (0.04) (0.09) (0.38) cutter raven’s score 1.96 2.00 1.50 (0.19) (0.21) (0.50) N 57 18 4 All-responder sample

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Correlates of Adoption, All-Responder Sample, Aug. 2013

Cash/No Drop Tech Drop Non-Adopter Adopter Non-Adopter Adopter log avg output/month 9.58 11.86 9.55 9.68 (0.17) (0.26) (0.80) log avg employment 3.48 7.69 3.25 3.33 (0.17) (0.22) (0.75) log avg price, size 5 promo 5.51 5.44 5.36 (0.04) (0.04) (0.07) log avg price, size 5 training 6.11 6.59 5.99 5.76 (0.05) (0.04) (0.10) avg % promotional (of size 5) 38.73 0.00 36.15 27.51 (3.47) (5.42) (9.34) avg profit %, size 5 training 0.08 0.09 0.09 0.08 (0.01) (0.01) (0.01) avg Rs/ball, head cutter 1.48 1.90 1.57 1.63 (0.04) (0.07) (0.34) N 73 1 27 5 Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Correlates of Adoption, Initial-Responder Sample, March 2014

Cash/No Drop Tech Drop Non-Adopter Adopter Non-Adopter Adopter log avg output/month 9.37 8.18 9.32 10.07 (0.21) (0.36) (0.59) log avg employment 3.31 2.44 3.16 3.29 (0.21) (0.33) (0.45) % standard (of size 5) 91.67 100.00 91.87 84.29 (1.76) (3.03) (14.12) CEO university indicator 0.39 0.00 0.43 0.50 (0.07) (0.14) (0.22) CEO experience 17.59 14.00 16.47 15.14 (1.53) (2.93) (5.04) age of firm 27.24 11.00 22.53 25.29 (2.87) (2.49) (4.78) log avg price, size 5 promo 5.49 5.52 5.45 5.33 (0.04) (0.07) (0.05) log avg price, size 5 training 6.06 6.05 5.98 5.85 (0.06) (0.06) (0.09) avg % promotional (of size 5) 41.07 40.00 40.27 37.79 (3.81) (7.92) (6.19) avg profit %, size 5 training 0.08 0.04 0.09 0.10 (0.01) (0.01) (0.01) head cutter experience 22.29 29.00 16.44 19.50 (2.86) (2.53) (0.50) head cutter tenure 11.67 3.00 12.38 11.50 (2.25) (2.28) (8.50) avg Rs/ball, head cutter 1.47 1.45 1.62 1.55 (0.04) (0.09) (0.24) cutter raven’s score 2.00 1.00 1.89 2.00 (0.19) (0.20) (0.58) N 56 1 15 7 All-responder sample

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Correlates of Adoption, All-Responder Sample, March 2014

Cash/No Drop Tech Drop Non-Adopter Adopter Non-Adopter Adopter log avg output/month 9.60 10.02 9.31 10.39 (0.17) (1.84) (0.25) (0.53) log avg employment 3.49 5.06 3.14 3.64 (0.17) (2.62) (0.25) (0.47) log avg price, size 5 promo 5.51 5.52 5.46 5.36 (0.04) (0.05) (0.04) log avg price, size 5 training 6.11 6.32 6.00 5.87 (0.06) (0.27) (0.05) (0.07) avg % promotional (of size 5) 38.72 20.00 34.29 31.32 (3.52) (20.00) (6.18) (6.54) avg profit %, size 5 training 0.08 0.06 0.09 0.09 (0.01) (0.02) (0.01) (0.01) avg Rs/ball, head cutter 1.48 1.68 1.62 1.45 (0.04) (0.22) (0.08) (0.19) N 72 2 22 9 Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Adoption Regressions, Scale & Quality Variables

• Dep. var.: indicator for currently using offset die

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) tech drop group 0.14** 0.14** 0.17 0.16** (0.07) (0.07) (0.52) (0.07) cash drop group

• 0.02

(0.03) log avg output/month 0.07** 0.07** 0.03 0.06 (0.03) (0.03) (0.03) (0.04) log avg output*tech drop

• 0.00

(0.05) share standard (of size 5)

• 0.30
• 0.29
• 0.44

(0.28) (0.30) (0.28) log avg price, size 5 training

• 0.03
• 0.19*

(0.06) (0.11) avg share promotional (of size 5)

• 0.14**
• 0.15

(0.07) (0.10) avg profit rate, size 5 training 0.17 0.35 (0.47) (0.62) constant 0.03 0.03

• 0.47*
• 0.55**

0.34 0.11 0.23 0.13* 0.05 1.15* (0.05) (0.05) (0.27) (0.25) (0.29) (0.46) (0.34) (0.07) (0.05) (0.68) stratum dummies Y Y Y Y Y Y Y Y Y Y R-squared 0.13 0.13 0.13 0.22 0.14 0.16 0.06 0.08 0.07 0.36 N 106 106 106 106 77 77 93 105 88 65

◮ Full sample. Linear probability model.

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Adoption Regressions, Manager & .

• Dep. var.: indicator for using offset die (> 1000 balls) as of Aug. 2013

(1) (2) (3) (4) (5) (6) (7) (8) (9) tech drop group 0.14** 0.17** (0.07) (0.08) CEO university indicator 0.04 0.01 (0.07) (0.05) CEO experience (/100)

• 0.24
• 0.11

(0.17) (0.20) age of firm (/100)

• 0.04

0.02 (0.08) (0.11) Rs/ball, head cutter 0.05 0.08 (0.12) (0.15) head cutter experience (/100)

• 0.03

(0.09) head cutter tenure (/100)

• 0.19

(0.23) cutter raven’s score

• 0.01

(0.03) log avg output/month 0.04 (0.03) constant 0.03 0.05 0.11 0.07

• 0.02

0.00 0.03 0.03

• 0.41

(0.05) (0.05) (0.07) (0.06) (0.15) (0.01) (0.03) (0.07) (0.29) stratum dummies Y Y Y Y Y Y Y Y Y R-squared 0.13 0.09 0.09 0.06 0.07 0.12 0.13 0.18 0.26 N 106 71 78 98 98 34 33 37 65

◮ Full sample. Linear probability model.

Return

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Results, Incentive Payment Experiment, Die Purchase as Outcome

• Dep. var.: purchased first offset die (beyond trade-in) after Round 4 survey

All Strata Initial Non-Adopters First Stage OLS Reduced Form (ITT) IV (TOT) First Stage OLS Reduced Form (ITT) IV (TOT) (1) (2) (3) (4) (5) (6) (7) (8) rec’d treatment 0.42** 0.40** 0.40** 0.38** (0.15) (0.16) (0.16) (0.17) assigned to group A 0.68*** 0.27** 0.62*** 0.23* (0.12) (0.12) (0.14) (0.12) stratum dummies Y Y Y Y Y Y Y Y R-squared 0.57 0.40 0.24 0.40 0.50 0.40 0.22 0.40 N 31 31 31 31 26 26 26 26

◮ Regressions include stratum dummies.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Results, Incentive Payment Experiment, Current Use as Outcome

• Dep. var.: currently using offset die (have produced > 1000 balls with it)

All Strata Initial Non-Adopters First Stage OLS Reduced Form (ITT) IV (TOT) First Stage OLS Reduced Form (ITT) IV (TOT) (1) (2) (3) (4) (5) (6) (7) (8) rec’d treatment 0.39** 0.38** 0.48** 0.50** (0.16) (0.15) (0.19) (0.19) assigned to group A 0.68*** 0.26** 0.62*** 0.31** (0.12) (0.11) (0.14) (0.13) stratum dummies Y Y Y Y Y Y Y Y R-squared 0.57 0.66 0.59 0.66 0.50 0.43 0.26 0.43 N 31 31 31 31 26 26 26 26

◮ Regressions include stratum dummies.

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Wald Arithmetic

◮ Model Yi = α + βTi + εi Ti, Zi (instrument) binary. ◮ Estimators:

• βols = Y T=1 − Y T=0 = 5/10 − 0/21 = .5
• βiv = Y Z=1 − Y Z=0

T Z=1 − T Z=0 = 5/15 − 0/16 10/15 − 0/16 = .5

• βitt = Y Z=1 − Y Z=0 = 5/15 − 0/16 = .33
• βfs = T Z=1 − T Z=0 = 10/15 − 0/16 = .67

Return Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen

Introduction Technology Experiment I: Tech Drop Model Experiment II: Incentive Payment Conclusion

Reasons for Non-Adoption (internal consumption)

firm code insufficient

• rders

too busy doubt profitable waiting for

• thers to

prove value waiting for

• thers to

iron out kinks cutters unwilling printing problems

• ther

production issues

• ther

13 2 3 1 34 2 1 60 2 1 67 2 1 82 2 1 91 4 3 1 2 93 3 2 1 101 3 1 2 110 3 2 1 118 1 122 1 129 1 145 3 1 2 164 3 1 2 171 2 1 3 307 1 315 5 3 1 2 4 319 2 3 1 3

◮ Sample is round-4 respondents who have had die in their factory but are not currently using it.

Organizational Barriers to Technology Adoption Atkin, Chaudhry, Chaudry, Khandelwal & Verhoogen