Incentives, Inspection and Monitoring in Offshore Outsourcing of Services: Findings From Field Resea

1
Incentives, Inspection and Monitoring in Offshore
Outsourcing of Services Findings From Field
Research Ravi AronYing Liu
2
Work Types In The Survey

• Retail Banking
• FA
• Transaction Processing
• Credit Card Customer Support
• Corporate Banking
• Loan / Asset management
• Working Capital Analysis Cash Flow forecasts
• Investment Banking
• Equity Research
• Bond Pricing Research
• MA - Background Research
• Legal Services
• Patent Litigation Support
• Patent Overlap Analysis
• Pharmaceutical
• Bioinformatics
• Consumer Marketing
• Market Segmentation
• Consumer Behavior Analysis and Forecast

3
Survey Data Sources

• 2001 India 2 Surveys of Captive Centers
• 2001 Singapore 2 Surveys 1 BPO and 1 Captive
  Center
• 2002 Singapore 2 Surveys 2 BPOs
• 2002 Thailand 1 Survey JV
• 2002 India 3 Surveys 2 BPOs, 1 Captive
  Center
• 2003 Mauritius 4 Surveys 3 BPOs and 1
  Captive Center
• 2003 Singapore 3 Surveys 1 BPO and 1 Captive
  Center and 1 EOF
• 2003 India 4 Surveys 3 BPOs and 1 EOF
• 2004 India 4 Surveys 2 BPOs, 2 EOF
• 2004 Mauritius 2 Surveys 1 BPO and 1 Captive
  Center
• 2004 Singapore 2 Surveys 1 BPO and 1 EOF
• 2004 Thailand 1 Survey JV
• 2005 China 1 Survey BPO
• 2005 China 1 Survey Captive Center
• 2005 Singapore 3 Surveys - BPO
• 2005 India 6 Surveys - BPO
• 2005 India 2 Surveys Captive Center
• 2005 India 3 Surveys EOFs.
• 2006 India - 2 - Surveys - Captive Centers

4
Part I What Process Characteristics Affect
Quality Of Output?
An Exploratory Survey To Analyze The Nature
Extent of Operational Errors in
Information-Intensive Processes
5
Data Sources

• 2001 to 2003
• 5 firms and 33 processes for this project.
• For each process we took two observations 6
  months apart resulting in 66 observations in all.
  
• Regression Model

6
Codifiability And Verifiability

• Codifiability The extent to which it is possible
  to specify the set of agent responses to states
  of the world that can be encountered in executing
  a set of tasks (process).
• Verifiability The Ratio of the number measures
  of quality on which buyers and sellers express
  convergent understanding to the total Number of
  measures of quality as stated by the buyer.
• We term this ratio as the Objectivity Quotient
• Error Rate (Dependent Variable) Defined as the
  fraction of processes that were found to have
  errors in completed output (per measurement
  period).

7
Variables Description
8
Operationalization Of Variables Codification
9
Operationalization Of Variables Codification
10
Results
Standard error in parenthesis -plt.001, -
plt.01, - plt.05
11
Part II Effectiveness of Governance Instruments
12
Research Model
Contract Completeness
Codifiability (C)
Inspection Costs (IN)
Incentive Contracting
Output Quality
Incentive Multiple (IC)
Agents Incentive
Penalties Multiple (PE)
Provider Firms Monitoring Effort (PM)
Provider Firms Monitoring
Process-Specific System (PSS)
Buyer Firms Monitoring Effort (BM)
Buyer Firms Monitoring
Process Owner (OW)
13
Operationalization Of Variables Client
Provider Monitoring
Monitoring
14
Dataset Profile

• Surveys 2003-2004
• 11 firms, 80 processes
• Cross Sectional Data
• Regression Model

15
Standard error in parenthesis -plt.001, -
plt.01, - plt.05
16
Results

• Primary Factors
• Monitoring effort by client
• Monitoring effort by provider
• Process Codification
• Secondary Factors
• Process Owner within buyers firm (small effect
  size)
• Penalty built into contracts
• Insignificant Factors
• Number of agents per process cycle
• Incentives for quality

17
How to Solve The Causality Problem
18
Data Source

• Survey Two Isolate factors that vary from period
  to period (fortnightly) and see how quality
  varies over time
• Process volume
• Buyers effort
• Providers effort
• Time Series Data 2003-2007
• 4 firms, 33 processes - WIP
• 2 firms, 10 processes - completed
• 18 months of data
• 2 Measurements per month for each process
• 36 time stamps per process
• Regression model

19
Preliminary Results
Standard error in parenthesis -plt.001, -
plt.01, - plt.05
20
Comments

• There is strong support for the insight that
  monitoring by client is a highly effective means
  of driving quality outcomes.
• We wish to test if this is true across each
  process when we take into account process level
  and firm-level heterogeneities.
• Process level tests
• We run GLS models at the level of each process

21
GLS Results
22
The Benefit of Dual Monitoring
23
Relative Effectiveness Of Monitoring Efforts
Client Provider Compared
24
Summary Of Results

• Instruments of governance (monitoring and
  control) are in general more effective in
  inducing higher quality of output than
  contract-based instruments - incentives and
  penalties.
• Monitoring and Managerial Intervention
• Buyers (clients) managerial efforts are
  generally more effective than providers efforts
  in inducing higher quality
• Buyers managerial efforts are most valuable
  where processes are least codifiable.
• Extensions and Future Research
• Role of pricing
• Use of collaborative systems
• Use of captive centers to monitor and manage
  third-party BPOs and hybrid governance forms