Andrew McAfee - PowerPoint PPT Presentation

Loading...

PPT – Andrew McAfee PowerPoint presentation | free to download - id: 1650a-YzQ3O



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Andrew McAfee

Description:

Andrew McAfee. TOM Seminar October 9, 2003. Information Technology and Industry ... Relationship between IT intensity and industry concentration ... – PowerPoint PPT presentation

Number of Views:389
Avg rating:3.0/5.0
Slides: 36
Provided by: Roy9150
Learn more at: http://www.hbs.edu
Category:
Tags: mcafee | andrew | mcafee

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Andrew McAfee


1
Information Technology and Industry
Concentration Evidence and Explanations
  • Andrew McAfee
  • TOM Seminar October 9, 2003

2
Information Technology and Industry
Concentration Evidence and Explanations
  • Andrew McAfee
  • TOM Seminar October 9, 2003

3
The First Slide
(higher value greater concentration)
Low IT Industries
High IT Industries
4
The First Slide
(higher value greater concentration)
Low IT Industries
96 00 2.8
96 00 8.9
High IT Industries
5
Agenda
  • Relationship between IT intensity and industry
    concentration
  • Explanations for observed relationship
  • Theory on IT, concentration, and competition

6
Aggregate Concentration in the US
  • White (2002)
  • the data show clearly and strongly that -
    despite the recent waves of mergers - aggregate
    concentration declined during the 1980s and early
    1990s and then generally increased after the mid
    1990s
  • the size distribution of firms became slightly
    skewed toward larger firms
  • improved technologies of managing and
    monitoring may have helped overcome the inherent
    difficulties of managing larger enterprises

7
US IT Capital Stock
Tangible Wealth Non-inventory assets,
regardless of accounting treatment
8
IT Capital Stocks by Industry
9
IT Capital Stocks by Industry
Non-Depository Institutions
Wholesale Trade
Industrial Machinery
Brokers
Personal Services
All industries
10
IT Capital Stocks by Industry
11
IT Capital Stocks by Industry
Railroads
Agricultural Services
Coal Mines
Metal Mines
12
IT and other Tangible Wealth
Thousands of dollars per employee in
13
IT and Productivity
  • Stiroh (2002) Industry-level analysis
  • High IT industries experienced greater
    productivity gains after 1995 than did low IT
    industries
  • these results establish a link between IT
    capital and subsequent productivity growth across
    U.S. Industries.
  • Industry-level IT data from US BEA Tangible
    Wealth Survey

14
Research Question IT and Concentration
  • What happens to an industrys structure as its
    companies adopt more IT?
  • Greater concentration?
  • Less concentration (greater competitiveness)?
  • No significant change?
  • Differences across time periods?
  • Differences across industries?

15
Independent and Dependent Variables
  • Industry Concentration Compustat
  • Herfindahl Index
  • Does this industry have a few big firms, or a lot
    of small ones?
  • For all companies (mkt. Share1)2 (mkt.
    Share2)2
  • Monopoly 10,000
  • 100 equal size firms 100
  • Higher HI more concentration
  • Also Cx measures
  • of market share held by top x firms
  • IT Intensity BEA Tangible Wealth Survey
  • US economy 62 industries
  • Reduced to 58
  • 60 spending categories
  • Stocks and flows
  • 12 IT spending categories
  • Through 2000

16
Model 1 Discrete Measure of IT Intensity
  • HI Herfindahl Index
  • Dummy D 1 if year is post-95
  • Dummy C 1 if industry is IT intensive
  • IT measure of total tangible wealth
  • Threshold median across all industries
  • d differential acceleration of IT intensive
    industries after 1995

17
Model 1 Results
Dependent variable dln(HI), time period
1983-2000
Models are OLS, corrected for and
intercorrelation and heteroskedasticity across
panels
Significant at the 10 level Significant at
the 5 level
18
Model 1 Interpretation
  • Dependent variable change in HI
  • a 0.21, b 0.98, g -2.15, d 3.27
  • Pre-95 growth rates
  • Low IT a
  • High IT ag
  • Post-95 growth rate acceleration
  • Low IT b
  • High IT bd

19
Alternative Model Specifications
  • Dependent variables
  • Herfindahl Index, C4, C8
  • Level, change, d ln(HI)
  • Weights by industry size (square root of FTEs)
  • IT intensity threshold mean and median
  • Break year 1995, 1996
  • Time periods 1983-2000, 1990-2000
  • Industry fixed effects

20
Model 2 Controls and Continuous IT Intensity
Measure
  • IT95 Continuous IT intensity measure
  • of total industry tangible wealth in 1995
  • h late 90s concentration acceleration
    associated with 1995 IT intensity
  • Controls Intensity of other tangible wealth
    categories

21
Model 2 Results
Dependent variable dln(HI), time period
1983-2000
Significant at the 10 level, Significant at
the 5 level, Significant at the 1 level
22
Model 3 Concentration in late 90s
  • 1995-2000 time period
  • T year variable, 1-6
  • Dummy C 1 if industry is IT non-intensive

23
Model 3 Results
Dependent variable HI, time period 1996-2000
Significant at the 5 level, Significant
at the 1 level
24
Model 3 Results
Dependent variable HI, time period 1996-2000
Significant at the 5 level, Significant
at the 1 level
25
Future analyses
  • EBITDA as dependent variable
  • Services only, manufacturing only, etc.
  • Drop industries
  • FIRE outliers
  • IT-producing industries
  • Model 3 for different time periods
  • Additional controls?
  • What else?

26
Is this a Surprising Relationship?
  • No. IT as a Strategic Asset
  • Leveraging IT endowment
  • Yes. IT Doesnt Matter
  • Most IT endowments are easily replicable
  • Cheap, shrink-wrapped
  • IT as Strategic necessity
  • Yes. The Electronic Markets Hypothesis
  • IT leads to the dispersal of economic activity
  • Similarity
  • Concentration on exploitation of an IT endowment
  • What about acquisition?

27
An Acquisition-based Explanation for the IT
Industry Structure Relationship
  • Some applications are organizationally hard to
    adopt some are organizationally easy
  • The two types provide different benefits
  • In some industries, the benefits from hard IT
    justify the cost and risk in others, they do not
  • All industries have easy IT
  • High IT industries also have hard IT
  • Industries where hard IT matters concentrate over
    time as successful adopters gain market share at
    the expense of unsuccessful ones
  • Pattern is different in industries where easy IT
    alone suffices (i.e. low IT industries)

28
Easy vs. Hard IT
General Managers Point of View
IT Efforts Impose Transitions
  • Easy IT
  • Spontaneously accepted
  • Doesnt cause conflict
  • Esp. in advance
  • Isnt perceived as organizational change, threat
  • Improves productivity, performance
  • Email, Corporate website, Intranet, Powerpoint,
    etc.
  • Hard IT
  • Is (correctly) perceived as organizational change
  • Conflict starts with proposal
  • Requires decision-making, negotiating
  • New IT cannot easily be ignored
  • Embeds novel business processes
  • ERP, Supply Chain, CRM, B2B Exchange, etc.

29
Organizational change and hard IT
What changes?
Organization change brought by process-based hard
IT is
  • Tasks
  • Sequences
  • Possible branches
  • Decision rights
  • Roles
  • Status
  • Authority
  • Power
  • Centrality
  • Externally Imposed
  • Interdependence-increasing
  • Sudden
  • Large-scale
  • Core, often
  • Unignorable

30
So Why Attempt Hard IT?
  • Hard IT brings benefits
  • Process standardization and improvement
  • Compliance deployment
  • Optimization
  • Monitoring, analysis, reporting
  • Integration and extension
  • that are difficult to replicate
  • without IT
  • with only easy IT

31
Low vs. High IT Industries
  • Low IT industries
  • Easy IT is enough
  • Hard IT isnt worth it
  • Benefits are not worth the difficulty, cost,
    risk
  • Bases of competition include
  • Commodity prices
  • Efficiency of manual labor
  • Relationships / network
  • Physical Barriers to Entry
  • Examples
  • Mining
  • Auto repair
  • Farming
  • Real Estate
  • Pipelines
  • Farming
  • Utilities

32
Low vs. High IT Industries
  • Examples
  • Wholesale trade
  • Financial Services
  • Industrial Machinery
  • Electronics
  • Health Services
  • Printing and Publishing
  • Transportation Services
  • Apparel
  • High IT Industries
  • Easy IT is not enough
  • Must also invest in hard IT
  • Bases of Competition include
  • Structured multi-party Coordination
  • Execution of operational processes
  • Speed, accuracy, consistency, etc.

33
The mid 1990s
  • Hard IT became more available in mid 1990s
  • Packaged enterprise and extra-enterprise
    applications
  • Enterprise Resource Planning
  • Supply Chain Management
  • Customer Relationship Management
  • The Internet
  • Process span not limited by network availability,
    transmission cost

34
Future Work
  • IT penetration rates, investment levels
  • By application category (easy vs. hard)
  • By industry
  • By year
  • Articulation of easy vs. hard IT
  • Business process imposition
  • Fit?
  • Technical considerations?

35
IT Endowments
  • Within IT strategy literature
  • Exploitation
  • Selection, Fit
  • Acquisition
  • Expense, technical difficulties (coding)
  • From other IT literature
  • Technology Introduction organization change
  • Enterprise application implementation failure
    rates 30-75
  • McDonagh (2001) extant empirical research
    supports the assertion that economic and
    technical considerations are unlikely to feature
    prominently when IT fails to deliver,
  • Failure rates for email, word processing,
    business intelligence?
About PowerShow.com