Lecture 10 Decision Making and Hierarchies

1
Lecture 10 Decision Making and Hierarchies

• Module III Topic 1

2
Introduction
Economics of HRM
Decision Making and Hierarchies

• How can you make decisions?
• Average the predictions of many people
• Meet to deliberate and discuss the decision
• Use the help of a person who you prefer
• Use some form of a price system where people who
  are correct are rewarded
• Put the question on the internet and see how
  people respond

3
The Wisdom of the Crowd
Economics of HRM
Decision Making and Hierarchies

• The average of the crowd is often an excellent
  predictor
• The weight of a horse
• The amount of money in a jar
• The U.S. submarine Scorpion
• It works when people are more likely to be
  right than wrong (Condorcet Jury Theorem)
• It does not work when people are more likely to
  be wrong than right (conventional wisdom is
  incorrect)
• Surveys are good in this regard
• The problem is that there is no penalty for a
  wrong answer, so there is no sorting in a
  survey. You can do potentially better

4
Committees and Discussion
Economics of HRM
Decision Making and Hierarchies

• Gather information in a group discussion rather
  than averaging. There are serious problems with
  this approach
• If people have similar views, group discussion
  tends to lead to extreme results. Other views
  are crowded out.
• People are reluctant to present their view if
  they believe that it is in the minority.
• The majority will tend to disregard a minority
  view as being incorrect so that new information
  is ignored.
• It may therefore be better to ask views
  individually
• There are situations where groups are good
• when it takes several people to put together a
  solution
• the solution is seen immediately when it is
  suggested

5
Prediction Markets
Economics of HRM
Decision Making and Hierarchies

• Markets where participants trade in contracts
  whose payoff depends on unknown future events
• derives from the efficient markets hypothesis
• in a truly efficient prediction market, the
  market price will be the best predictor of the
  event
• does not require that all individuals in a market
  be rational, as long as the marginal trade in the
  market is motivated by rational traders
• a number of successes in these markets, both with
  regard to public events like presidential
  elections and within firms, have generated
  substantial interest
• The Iowa experiment, HP forecasting techniques
• Google
• These are modeled after market economies and the
  invisible hand.

6
The Iowa Experiment
Economics of HRM
Decision Making and Hierarchies

• The original Iowa experiment, run in 1988,
  allowed trade in a contract that would pay 2 1/2
  cents for each percentage point of the popular
  vote in the presidential election won by Bush,
  Dukakis or others. More recently, it has run
  markets based on the 2003 California
  gubernatorial election, the 2004 presidential
  election, the 2004 Democratic presidential
  nomination, how the Federal Reserve will alter
  the federal funds interest rate and in the 2008
  US presidential elections.
• Universities in other countries have also started
  running event markets about their own elections,
  like the Austrian Electronic Market run by the
  Vienna University of Technology or the University
  of British Columbia Election Stock Market that
  focuses on Canadian elections
• In the week leading up to the last 4 elections,
  these markets have predicted vote shares for the
  Democratic and Republican candidates with an
  average absolute error of around 1.5 percentage
  points. By comparison, over the same four
  elections, the final Gallup poll yielded
  forecasts that erred by 2.1 percentage points.

7
The Iowa Experiment
Economics of HRM
Decision Making and Hierarchies
8
1. Organizational Design of an Economy
Economics of HRM
Decision Making and Hierarchies

• Optimal organization of an economy was one of the
  most debated issue in 20th century
• Centralized economies vs. decentralized economies
• Back again with the financial crisis too much
  deregulation?
• Leontief central planning (centralization) is
  efficient
• coordination, economies of scale, control
• Adam Smith metaphor for decentralized economies
• he intends only his own gain, is led by an
  invisible hand to promote an end which was no
  part of his intention By pursuing his own
  interest he frequently promotes that of society
  more effectually than when he really intends to
  promote it. (1776)
• Power of market economies to create economic
  value without the government playing a major role

9
Organizational Design of an Economy
Economics of HRM
Decision Making and Hierarchies

• Hayek market (decentralization) is more
  efficient
• Idea that markets are a form of collective
  intelligence a powerful information system that
  cannot be replicated by a central planner
• costly to move all info. to central planner
  decentralization makes better use of specific
  knowledge of time place
• How can fragments of knowledge existing in
  different minds bring about results which, if
  they were to be brought about deliberately, would
  require a knowledge on the part of the directing
  mind which no single person can possess?

10
Markets as Information Incentive Systems
Economics of HRM
Decision Making and Hierarchies

• Examples of markets as forms of organization
• prediction markets (insurance, financial, etc.)
• Market economies have 3 important features
• Markets make better use of local knowledge
  through decentralization
• prices provide substantial coordination, without
  central planning
• Because pricemarginal value of the good in its
  best alternative use
• Markets provide strong incentives for efficient
  resource allocation
• motivates good decision making
• moves decision rights to person with most
  valuable/ relevant specific knowledge
• motivates investments in human capital
• motivates creativity / innovation

11
Organizational Design of a Firm
Economics of HRM
Decision Making and Hierarchies

• Organization design must address the same
  problems
• use local knowledge for effective decision making
• move the knowledge to the decision maker, or the
  decision to the knowledge
• coordinate across the firm
• provide good incentives for both
• Keep innovation adaptation
• To do these, we need to think about
• centralization v. decentralization
• key pieces of knowledge that drive your firms
  business
• finding developing talent
• performance evaluation incentives
• Can we design an organization to mimic a market?
• even if we cant completely, the intuition is
  very useful

12
2. Centralization vs. Decentralization
Economics of HRM
Decision Making and Hierarchies

• Q How do you allocate a decision?
• First, break the decision into stages e.g.,
• initiatives (brainstorming)
• ratification (strategy)
• implementation (tactics)
• Monitoring (control)
• Next ask, Who has the relevant knowledge to make
  the decision at that stage?
• two kinds of knowledge
• needed to make the decision itself
• of how the decision affects others for
  coordination

13
Specific Knowledge
Economics of HRM
Decision Making and Hierarchies

• Hayek emphasized specific knowledge of time
  place
• local knowledge needed to make decisions, that is
  costly to communicate to a Central Planner,
  thus likely to be ignored under centralization
• knowledge lies on a spectrum from more costly
  (specific) to communicate to less costly (general)

14
Specific Knowledge
Economics of HRM
Decision Making and Hierarchies

• Attributes of knowledge/info that make it more
  specific
• costly to transfer
• perishable
• complex
• costly to understand
• requiring scientific or specialized technical
  skills
• subjective or experiential
• unreliable / risky to use
• noisy (garbling), you need to communicate
  frequently so it is costly

15
Benefits of Decentralization
Economics of HRM
Decision Making and Hierarchies

• Specific knowledge favors decentralization
• Better use of specific knowledge dispersed
  throughout the organization
• When valuable knowledge is costly to communicate,
  consider decentralizing the decisions
• creativity is an important example
• Prevents senior management from being overwhelmed
• Training/ development intrinsic motivation for
  lower level managers
• Less bureaucratic/ more manageable scale
• Otherwise, there are important benefits to
  centralization
• As in all things, balance this tradeoff

16
Benefits of Centralization
Economics of HRM
Decision Making and Hierarchies

• Consider centralizing the decision if specific
  knowledge is not
• crucial, or if market failures inside the firm
  are important
• Economies of scale (natural monopoly)
• Common assets physical capital, brand name
  reputation
• Aggregate knowledge from across the organization
• experience of the combined organization
• Coordination (externalities)
• synchronization, consistency
• cooperation instead of conflict
• Standardization
• control
• common strategy

17
3. Decision Making, Hierarchy, Control
Economics of HRM
Decision Making and Hierarchies

• Decision making as a 4-stage process
• 1. initiatives
• 2. ratification
• 3. implementation
• 4. monitoring
• Different stages can be more centralized or
  decentralized

18
Creativity Control in Decision Making
Economics of HRM
Decision Making and Hierarchies

• By breaking a decision into stages, you can
  allocate them to different levels
• this gives benefits of decentralization
  centralization at the same time
• steps 1 3 usually need lower-level specific
  knowledge decentralize
• these are the creative steps
• steps 2 4 usually need coordination w/ the rest
  of the org centralize
• these are the control steps
• There is a fundamental tradeoff between
  creativity control
• for the same resources, more creativity implies
  less control, vice versa
• lets look at a simple example to see why

19
How Much Decision Control?
Economics of HRM
Decision Making and Hierarchies

• Consider 2 firms with 2 employees Gladys and
  Willie
• Firm imports womens lingerie and sleepwear
• Young and funky image
• Gladys has to decide whether to branch out into
  more romantic lingerie
• Upfront investment in marketing, distribution and
  production
• Can be worth it but can also be a disaster
• Two types of errors
• False positive invest in the line and it turns
  out unprofitable
• False negative not invest while it would have
  been profitable

20
How Much Decision Control?
Economics of HRM
Decision Making and Hierarchies

• Different structures
• The units evaluate new ideas differently
• Hierarchy W evaluates new ideas, passes some
  to G. G approves or rejects those
• Flat GW both evaluate new ideas, separately
• N of ideas each can evaluate per period
• flat firm evaluates twice as many ideas per period

Flat
Hierarchy
21
Notation
Economics of HRM
Decision Making and Hierarchies

• N of ideas a creative worker generates each
  period
• the Flat structure generates twice as many new
  ideas per period
• p probability of correct decision at 1st
  evaluation
• p gt ½, or the firm should toss a coin to make
  decisions
• q probability of correct decision at 2nd
  evaluation
• q gt p is a reasonable assumption, but not
  important
• There are two kinds of mistakes to worry about
• False Positive implementing a bad idea
• False Negative rejecting a good idea

22
Hierarchy
Economics of HRM
Decision Making and Hierarchies
Good
New Idea
Bad
23
Flat
Economics of HRM
Decision Making and Hierarchies
Accepts p
Good
Rejects (1-p)
New Idea
Accepts (1-p)
Bad
Rejects p
Gladys or Willie Evaluates
24
Results
Economics of HRM
Decision Making and Hierarchies
25
Are Hierarchies Conservative?
Economics of HRM
Decision Making and Hierarchies

• Flat structures
• evaluate ideas more quickly
• evaluate more ideas for the same of employees
• make more changes, good bad
• have more successes failures
• Environments favoring a more hierarchical or
  flat structure?

26
Authority Structure and Errors
Economics of HRM
Decision Making and Hierarchies

• Hierarchical
• Reduce false positive and increase false negative
• Approve fewer projects overall
• Good where careful consideration is needed
• Good with traditional industry regulated
  industry
• Bad for rapid change
• Ex. Exxon Valdez

Small Upside, Large Downside
27
Authority Structure and Errors
Economics of HRM
Decision Making and Hierarchies

• Flat
• Reduce false negative and increase false positive
• Creative people not attracted to hierarchical
  firm
• Good when unprofitable projects are not too
  costly or when profitable projects are likely to
  be very profitable
• Ex. startup

Large Upside, Small Downside
28
Trading off Creativity Control
Economics of HRM
Decision Making and Hierarchies

• A cost of decentralization is loss of control a
  benefit is creativity
• For the same resources
• improving control inevitably implies that
  creativity suffers, vice versa
• There is only one way out of this tradeoff
  spending more resources to increase accuracy of
  decisions
• hire more skilled decision makers
• provide more training
• better information or analysis tools
• Other ways to tip the balance toward creativity
  or control
• personality of staff conservative or liberal
• incentives downside punishments upside
  rewards
• constraints on budgets to limit downside risk
  (e.g., budgets)

29
De(Centralization) Managing Change
Economics of HRM
Decision Making and Hierarchies

• Centralization applies more for large scale
  changes that need to be enacted quickly
• strategic changes in direction
• changes affecting multiple units
• Decentralization applies more for incremental
  changes
• implementation of strategic changes
• refinement of strategy for individual units
• gradual evolution

30
4. How To Structure a Hierarchy?
Economics of HRM
Decision Making and Hierarchies

• Firms face a tradeoff in how to structure a
  hierarchy
• A greater number of levels increases costs in
  several ways, in addition to less innovation
• Information has to be passed between more
  managers
• Information processing and decision making takes
  longer, as each step of communication takes a
  certain amount of time
• If the firm wishes to reduce the number of
  levels, it must expand the number of managers
  within each level in order to perform the same
  amount of work.
• That would be a flatter structure
• A flatter structure has a greater span of
  control, or number of employees reporting to each
  manager
• A steeper structure has a lower span of control.

31
Type of Hierarchies
Economics of HRM
Decision Making and Hierarchies
32
Span of Control and Hierarchy Levels
Economics of HRM
Decision Making and Hierarchies

• A flatter structure reduces the costs implied by
  adding layers to the hierarchy, but creates its
  own costs
• Each manager must now supervise and direct more
  subordinates
• Takes more of the managers time, reducing the
  time that can be spent on other tasks
• Reduce the effectiveness at supervision, since
  attention is spread more thinly
• A hierarchy trades off span of control versus
  of levels
• Many factors affect the optimal span of control
  and number of hierarchical levels
• A lower cost of supervision will imply a larger
  optimal span of control, since the marginal cost
  of increasing the span of control by another
  subordinate is lower
• A lower cost of acquiring or passing on
  information will increase the optimal number of
  layers in the hierarchy, since it lowers the
  marginal cost of adding a new layer

33
A. Type of Tasks to be Performed
Economics of HRM
Decision Making and Hierarchies

• Routine tasks
• easier for the manager to supervise each employee
• manager has to spend less time training the
  subordinate, and deciding what to do in
  non-standard situations
• span of control will tend to be larger, and the
  number of hierarchical layers smaller.
• Complex tasks
• requires more input from the supervisor
• subordinates more likely to need the experience
  and skills of the supervisor in order to analyze
  the situation and decide what to do
• make impossible standard operating procedures
• smaller optimal span of control, and a greater
  number of levels in the hierarchy

34
B. Skills of Managers Subordinates
Economics of HRM
Decision Making and Hierarchies

• Higher skills of both managers and subordinates
  will tend to increase the optimal span of
  control, and reduce the number of layers of a
  hierarchy
• manager and subordinate can process more
  information and solve more difficult problems
• talented managers more effective at supervision
  and direction
• talented subordinates more effective at
  implementing the directions of their boss
• Since almost all firms put more talented managers
  in higher levels, this effect increases the
  optimal span of control at higher levels compared
  to lower levels

35
C. Incentive Mechanisms
Economics of HRM
Decision Making and Hierarchies

• To reduce agency problems, firms may
• rely on performance evaluation and incentive
  mechanisms
• monitor closely workers in order to detect
  shirking
• if performance evaluation is more effective, less
  monitoring is necessary, freeing up some of the
  supervisors time
• Firms having access to better incentive plans
  will have a larger span of control
• performance evaluation tends to be better the
  closer the manager is to the top of the hierarchy
• managers actions have a more direct effect on
  firm value
• better reflected in available performance
  measures
• span of control bigger at higher levels of the
  hierarchy

36
D. Acquiring Communicating Knowledge
Economics of HRM
Decision Making and Hierarchies

• Lowering the costs of acquiring knowledge
• increases the productivity of knowledge workers
  in a hierarchy
• Increases managers span of control
• modern information technology tends to be a
  complement to the work of managers in a
  hierarchy, rather than a substitute
• raises the productivity of managers, and
  especially those who are highly skilled and at
  higher levels
• Lowering the costs of communication
• similar effects
• managers and subordinates are able to communicate
  with each other more effectively, cheaply, and
  quickly
• allows the manager to oversee more subordinates
• the effects of advances in information technology
  seems to increase the span of control, leading to
  flatter organizations (Rajan and Wulf, 2006)

37
Acquiring Communicating Knowledge
Economics of HRM
Decision Making and Hierarchies

• Advances in information technology has an
  ambiguous effect on the number of layers
• Lower cost of communication tends to increase the
  optimal number of layers, as communication is
  faster
• On the other hand, the greater productivity of
  knowledge workers, combined with higher spans of
  control, implies that more can be accomplished by
  each layer of the hierarchy, implying that fewer
  layers are needed for a given level of output.

38
5. Key Ideas
Economics of HRM
Decision Making and Hierarchies

• The knowledge problem of organizational design
• specific versus general knowledge
• coordination types mechanisms
• Decision making
• decentralization v. centralization
• decision management v. control
• degrees of decision control / hierarchy, their
  effects
• Two types of errors in decision making
• tradeoff between creativity control
• Hierarchical levels and span of control
• cost of monitoring workers costs of acquiring
  and passing information