PRI Institutions and Education

1
PRI Institutions and Education

• Lant Pritchett
• Conference on Human Development in India

2
Two parts

• Suppose a state had decided to decentralize
  responsibility for education to the PRIwhich
  level should do what?
• Handling teachersthe elephant in the elephant

3
How does on allocate responsibilities across the
tiers of PRI?

• First, unbundle by activity
• Second, what are public finance first
  principles that apply?
• Third, what are the accountability first
  principles that apply?

4
Rough sizes of jurisdictions
Level People Schools
State 30-80 million
District 800,000-2 million 400-thousands
Block 60,000- 600,000 40-300
GP Few thousands A few to many
5
First principles of public finance
Economies of scale Externalities Equity Heterogeneity of demand
Standards State State State State
Planning District/GP District State
Asset creation GP None State GP
Operation School None State School
Monitoring and Evaluation State State State State
6
(No Transcript)
7
First principles of accountability
Discretionary? Transaction intensive? Who can best infer performance?
Standards No No Technical
Planning A bit A bit A bit technical
Asset creation Yes Yes Local
Operation Yes Yes Local
Monitoring and Evaluation No Yes Technical
8
Allocations of responsibility
My ideal based on first principles De jure De facto
Standards State State Little to none
Planning District State/district State/district
Asset creation GP (support from district) District/GP District/GP
Operation School/GP (with district) State State
Monitoring and Evaluation State State Little to none
9
Diagnosis

• In spite of highly paid teachers the public
  system has
• High teacher absenteeism
• Continued low access
• Low levels of learning achievement
• Losing students to private education

10
Nearly every state is worse than any other
country
11
Still have not reached primary completion
12
Low levels of learning acheivement
Table 7 Estimate of the fraction of the with inadequate learning achievement in mathematicseither drop out below grade 5 or score less than 50 percent in grade 5in the four states Table 7 Estimate of the fraction of the with inadequate learning achievement in mathematicseither drop out below grade 5 or score less than 50 percent in grade 5in the four states Table 7 Estimate of the fraction of the with inadequate learning achievement in mathematicseither drop out below grade 5 or score less than 50 percent in grade 5in the four states Table 7 Estimate of the fraction of the with inadequate learning achievement in mathematicseither drop out below grade 5 or score less than 50 percent in grade 5in the four states Table 7 Estimate of the fraction of the with inadequate learning achievement in mathematicseither drop out below grade 5 or score less than 50 percent in grade 5in the four states Table 7 Estimate of the fraction of the with inadequate learning achievement in mathematicseither drop out below grade 5 or score less than 50 percent in grade 5in the four states Table 7 Estimate of the fraction of the with inadequate learning achievement in mathematicseither drop out below grade 5 or score less than 50 percent in grade 5in the four states
State Estimate of fraction of children with inadequate primary learning achievement (sum of columns II and V) Percent reporting not having completed grade 5 (II) Percent with less than X percent correct (NCERT assessment in 2002) Percent with less than X percent correct (NCERT assessment in 2002) Percent with less than X percent correct (NCERT assessment in 2002) Percent with less than X percent correct (NCERT assessment in 2002)
State Estimate of fraction of children with inadequate primary learning achievement (sum of columns II and V) Percent reporting not having completed grade 5 (II) 30 (III) 40 (IV) 50 (V) 60 (VI)
Karnataka 92.3 34.90 22.6 38.8 57.4 74.4
Kerala 83.0 2.8 36.1 59.7 80.2 92.6
Rajastan 94.5 43.3 17.6 32.6 51.2 69.5
West Bengal 79.2 47.0 8.5 18.0 32.2 49.8
Source Singh, Jain, Gautam and Kumar 2004. Source Singh, Jain, Gautam and Kumar 2004. Source Singh, Jain, Gautam and Kumar 2004. Source Singh, Jain, Gautam and Kumar 2004. Source Singh, Jain, Gautam and Kumar 2004. Source Singh, Jain, Gautam and Kumar 2004. Source Singh, Jain, Gautam and Kumar 2004.
13
The elephant in the elephant Teachers

• Highly paid
• Zero accountability
• Politically powerful

14
Teacher salaries are high relative to GDP per
capita
15
What do we learn from the experience with
para-teachers

• West Bengal, Rajasthan, Madhya Pradesh have had
  substantial para-teacher experiences
• Some would argue that these show that
• changing teacher motivation and a little
  training means one can get the same or better
  performance for a fraction of the cost

16
Parental satisfaction in WB
17
Learning achievement in Rajasthan
18
Why does this work?

• Zero accountability means pay is not motivation
  it is a sinecure
• State cadre teachers dont want to be there
• Renewable contracts motivate
• A modest amount of relevant training can do a
  great deal of good at primary

19
Autonomy for accountability

• Lowest levels get cash plus responsibility with
  increased levels of internal and external
  accountability
• Internal accountabilitygreater separation of
  regulator function (higher levels) from
  provider functions means greater monitoring
  of --processes
• --outcomes
• External accountabilitybenchmarked, relevant
  information on inputs and outputs
• Strengthen inclusiveness and effectiveness of
  village le
• Technical is provided

20
A modest proposal

• The existing state employee teachers becoming a
  cadre that will phase out.
• All new teachers are hired into a new employment
  status as district employees
• The new scheme has three essential and new
  features.
• District hires/GP control assignments
• Career progression has two bumps/jumps with
  tripartite inputs

21
New professional cadre of teachers

• District hires teachers based on merit and
  recommendation criteria into a pool that are
  eligible for assignment to any school.
• Each GP/school controls who is actually assigned
  to their school from the eligible pool.
• What happens to teachers who are unassigned?
• What happens in schools that cannot fill posts?

22
New professional cadre of teachers

• Career of a teacher has three phasesapprentice,
  journeyman, master
• During apprentice period on a series of
  renewable at will contracts with the district
  and get paid only if assigned
• To make journeyman requires assessment of X
  years of performance by
• GP/school
• District teaching support
• Peer committee of master teachers
• (equivalent of making partner or tenure)

23
A new professional cadre

• Associates
• make a much higher salary
• have greater employment security
• Have regular merit and/or seniority pay
  increases
• But assignment still controlled by GP/school

24
Structure of professional cadre compensation
Master/senior/full
Jump to master rare and controlled, most spend
career as associates
Compensation
BIG jumps across levels
Associate/partner
Years of service
25
A new professional cadre

• Possibility of making one more career jump to
  master
• Again, much higher pay
• But much higher responsibilities (e.g. district
  wide, quality enhancement)
• Rarenot expected as a matter of course
• Based on district, peer, GP/school inputs

26
Transition

• Existing system finance and functionaries are
  disarticulated with massive accountability
  failure as a result (that is, mix of cash and
  in kind resources reach the school)
• New system all cash goes directly to the GP (with
  tax off the top for state, district overheads)
• Massive potential fiscal savings in realigning to
  a new professional teacher cadre implies improved
  access and improved quality is possible

27
Transition

• During transition GPs receive mix of cash and in
  kind.
• All new hires are in new system and each old
  cadre teacher is replaced with cash grantnot in
  kind entitlementthat is new teacher cost plus
• Eventually GP assumes full cash implications of
  each teacher it assigns (with implications for
  disadvantaged premia and journeyman and
  master premia)

28
(No Transcript)