International services outsourcing and innovation: an empirical investigation

1
International services outsourcing and
innovation an empirical investigation Holger
Görg and Aoife Hanley University of Nottingham
2
Why look at services outsourcing?

• Fear factor
• Media attention spurred economists to look into
  effects of services offshoring (e.g. Amiti and
  Wei (2006))
• But maybe a good thing economic restructuring?
• Reverse the hollowing out of industrial
  production ever shifting PPFs through increased
  innovation sustain a countrys competitive
  advantage

3
Our research question

• We empirically analyse the link between services
  offshoring and innovation at establishment level
• Does offshoring of services raise innovation?
• Data for ROI Amiti and Wei (2006) suggest
  services more important in ROI than in US
• We use data for both services and manufacturing
  sectors
• Time period 2000 - 2004

4
Background

• Empirical work on domestic market effects
  (Feenstra and Hanson, 1999 Head and Ries, 2002
  Hijzen et al., 2005)
• Theory Wage savings prompt North to o/s
  relatively less advanced intermediates from South
  (Glass and Saggi, 2001)
• Prediction Northern wages (relative to the
  South) fall due to outsourcing
• What makes Glass-Saggi special dynamic model
  with reinvestment of profits

5
Predictions of G-S model

• G-S and earlier work allow 2 effects to be
  identified
• Offshoring negative effect on wage rate
• Offshoring raises innovation in offshoring firm
• Non G-S effects
• Restructuring
• Learning of new technologies

6
Services offshoring in practice (1)

• Positive impact of cheaper transport,
  telecommunications improvements, cheaper
  computing and liberalisation (Gage and Lesher,
  2005)
• Overall importance of service inputs small (Amiti
  and Wei, 2006)
• Compelling cost reasons underscoring future
  growth
• 54 Germany
• 44 US
• 9.24 Russia
• 14 Portugal and China
• India

7
Services offshoring in practice (2)

• Services industry no longer means strictly
  services (Gage and Lesher, 2005)
• Some trade patterns in services violate textbook
  predictions e.g. computer programming from
  Hyderabad in India

8
Methodology (1)

• Innovation measured as RD intensity
• X contains other plant characteristics (lagged)
• Time and industry dummies
• Error terms µ (plant specific) and e (time
  specific)
• To test Glass-Saggi within transformation i.e.
  coefficients determined using within plant
  deviation from the mean in variables

9
Methodology (2)

• This equation allows the wage channel (from
  Glass-Saggi) to work
• Expected sign on ? 1t to be negative
• Estimate 1) and 2 separately with outs as
  exogenous
• Also try instrumenting outs in both 1) and 2)
• If e and f are correlated, we estimate using 3SLS

10
Data description

• Data from ABSEI collected by Forfás for 2000 to
  2004
• Survey of plants with at least 10 employees
• 55 60 percent response rate
• Spending on RD in Ireland only 0.8 of GDP
  (OECD, 1998)
• This means Ireland is low on the list of
  industrialised countries in terms of RD spending

11
Key variables in the data

Table 1
12
Innovation rates for firms that offshore services

Table 2
13
Which industries offshore services?

Table 3
14
Econometric results (1)

15
Econometric results (2)

16
Econometric results (3)

17
Econometric results (4)

18
Econometric results (5)

19
Conclusions

• Recent developments in Web and other media
  improve scope for offshoring service inputs
• Firm should increase innovation with increases in
  offshored services through the mechanism of wages
• Our results somewhat agree with G-S (ive
  relationship between offshored services and RD)
• This only holds for manufacturing sector firms
• This positive effect does not work through the
  wage channel, contrary to G-S