Outsourcing, Productivity, and Input Composition at the Plant Level

1. Outsourcing, Productivity, and Input Composition at the Plant Level Catherine J. Morrison Paul and Mahmut Yaşar World Congress, Washington D.C., May 2008

2. Make-or-Buy • “It is the maxim of every prudent master of a family, never to attempt to make at home what it will cost him more to make than to buy.” Smith (1776, pp.485-486) • “… quite apart from the question of diminishing returns the costs of organizing certain transactions within the firm may be greater than the costs of carrying out the exchange transactions in the open market.” Coase (1937, pp. 7-8)

3. Productivity drivers • Firms’ productivity depends on their potential to minimize production costs by substituting among a variety of inputs • Studies used to focus on “value added” capital and labor inputs but now recognize substitution of processed (materials) for processing (capital and labor) inputs • More generally, input choices include purchasing rather than producing goods and services (implies outsourcing or subcontracting) • Affects the productivity & competitiveness of outsourcing firms

4. Outsourcing drivers • Firms would be expected to use their own inputs for activities they have a comparative advantage for and outsource others • Roodhooft and Warlop (1999) • A firm’s comparative advantage depends on trade, price, and technical conditions • E.g., import competition from lower wage countries encourages foreign outsourcing, reducing the wages of less skilled laborers • Feenstra and Hanson (1999, 2003), Hanson et al. (2005) • In reverse, lower wage countries may subcontract output • Both specialized tasks and lower productivity/skill processes may be outsourced to other domestic or foreign firms

5. Our goal • We explore how outsourcing or subcontracting is related to plant productivity and input composition/substitution • from the perspective of a developing country • We consider the productivity effects of: • domestic outsourcing, through the share of inputs that are subcontracted • receiving contracts, through the share of output that is subcontracted • foreign outsourcing, through the share of imported intermediate inputs • represented as production frontier shifts and twists • from diff shares of subcontracted inputs/outputs or imported materials • and find this aspect of input choice important to recognize

6. Literature • Outsourcing/subcontracting has not been a focus of the production literature • But some studies suggest (domestic or foreign) outsourcing results if producing inputs or services in-house is at least as costly as subcontracting them • Involves two types of costs • production costs: labor costs, scale economies, clustering of special skills or expertise, or production smoothing • transactions costs: negotiating, monitoring, and enforcing contracts, or searching for outside suppliers • Also may be benefits from foreign technologytransfer through outsourcing • importing: R&D embodied in materials/capital imports and learning from their use (input subcontracting through input imports) • FDI/exports: increased demand for exports from host country suppliers (output subcontracting)

7. Data: output and inputs • Plant data collected by the State of Statistics in Turkey for the Annual Surveys of Manufacturing Industries • International Standard Industrial Classification, Apparel & Textiles • Output (Y): total value of shipments plus changes in inventories of finished goods and work-in-progress • Material (M): expenditures, allowing for changes in stocks • Energy (E): value of fuel+electricity • Y,M and E nominal values are divided by corresponding price deflators to get constant value quantities at 1987 prices • Labor input (LT and LN): technical and administrative workers, production workers • Capital (K): perpetual inventory method using gross investment data • Seven regions distinguished

8. Data: outsourcing and technology transfer • Our domestic outsourcing variables are: • subcontracted input (the share of a plant’s inputs subcontracted to supplier plants, in total inputs) • subcontracted output (the share of output subcontracted by other plants, in total output) • Klicaslan and Taymaz (2005) • Our foreign outsourcing variable is: • the share of imported materials in total materials use • Our foreign technology transfervariables include: • the share of exports in total sales • the share of foreign firm ownership (FDI)

10. Empirical framework • First look at premia/correlations • for Pit = performance indicators(t=time and i=plant) • labor productivity (LProd), wage, employment, skilled labor, and capital intensity • O/Sjit(j=SUBI, SUBO, IMPM) • D=regional (r) and time (t) dummies and EMP=size • Estimate:

12. Econometric Framework • We assume that the plant production function Y = f(X,r,t) can be represented by a translog approximation: • where Xj ( j = K,LT,LN,E,M) is the jth input, the rm include the O/S, technology transfer, characteristics and dummy variables distinguishing the plants, and  is a error term • Use Olley and Pakes (OP) method to estimate

14. SUBI Results • βSUBI is positive and significant (at 10%) • positive first order productivity effect of SUBI • βLT,SUBI is positive and significant (at 1%) • a strong positive LT bias • a higher share of subcontracted inputs implies greater administrative and technical (skilled) labor intensity • βLN,SUBI is significantly negative (at 5%) • more SUBI is associated with less use of unskilled labor • SUBI substantively affects only the labor shares • all other cross-effects are insignificant • The positive SUBIoutput elasticity is driven entirely by the positive first-order and LT effects

15. Interpretation • SUBI-LT complementarity and SUBI-LN substitutability • implies that subcontracting inputs reduces in-house “production” of lower-productivity and -skill activities or processes or services • e.g., those provided by unskilled production line workers or janitorial workers • rather than services of skilled workers • e.g., legal or accounting services • Perhaps because input subcontracting requires administrative and technical expertise due to greater required supervision or support of contractors.

16. SUBO Results • By contrast, more SUBOimplies significantly less reliance on skilled labor (LT) • implies subcontract-receiving plants are more “low-tech” • SUBO is also associated with lower M levels • suggests more use of lower value primary rather than processed materials • but a greater energy share • perhaps required for the additional processing • The lower M and LT levels for subcontract-receiving plants underlie the negative SUBO output elasticity.

17. IMPM Results • More IMPM also implies lower (but not sig) domestic materials and skilled labor use • The only sig IMPM bias is a positive E effect • The dominant effect of IMPM is therefore from the first order coefficient bIMPM • imports have little effect on input composition • the primarily negative input biases are too small to counteract the strong positive overall effect

18. Tech Trans Results, • Linkages of trade factors with O/S vary by mechanism • no significant productive relationships between SUBI and EXP, FDI, IMPM • but effects between SUBO and trade variables are sig • with EXP is negative, and with FDI and IMPM positive • IMP or FDI firms do not have as low productivity as other SUBO plants, but EXP firms actually have worse productivity • conversely, the productive contributions of trade factors are smaller for EXP and larger for IMP or FDI plants with SUBO • a higher EXP share is associated with a lower LN share • supports that internationally best practice production requires more skilled and less unskilled workers