Working with state governments
73. DFID currently has four 'focus states': AP, WB,
Orissa and MP. The Department explained their rationale for engaging
with individual states in terms of the greater impact which their
work can have at state-level.[124]
This is both because state governments are responsible for the
implementation of most development-related policies, and because
many of the critical constraints to development lie at state level.
DFID works through different strategies at different levels of
government. As far as we could see, DFID's work at state-level
consists of influencing the mechanisms for programme delivery,
monitoring programme implementation and facilitating coordination,
rather than through policy influence. DFID told us that there
was scope for them to engage in policy dialogue at state level,
but we were not shown any direct evidence of DFID's impact on
state-level policy development.
74. DFID's original selection criteria for their
four target states were: high levels of poverty, good state governance,
and the history of states' previous relationships with DFID (as
a proxy for DFID's comparative advantage).[125]
During our visit to India we saw evidence of the success of DFID's
'flagship' state, AP. As discussed in Chapter 1, AP does not have
particularly high levels of poverty. Instead it seems that AP's
reform-minded government and good governance were key factors
in DFID's decision to make the State one of their focus states.
In 1999, the GoAP launched a strategy paper, entitled 'Vision
2020'.[126] This envisaged
that together, the rapid development of the industrial and service
sectors of AP's economy, the modernisation of agriculture and
the development of a state-of-the art governmental sector, would
lead to the total eradication of poverty in the State. Since 1999,
Vision 2020 has been implemented through a diverse range of government
sector strategies, including poverty reduction, e-governance and
power sector reforms (discussed in paragraphs 58-60).
75. DFID chose AP as a focus state in 1999 on the
basis that they would be able to have a meaningful dialogue with
the State's government, and be able to encourage the processes
of reform which were already occurring under Vision 2020.[127]
To this end, between 1999 and 2004, DFID invested over a third
of its total India programme spend (a total of £248m.) in
AP (see table 1). Together with the World Bank, DFID has been
the principal advocate of AP's government reforms and market-led
development. DFID has supported high profile reform initiatives
including the expansion of e-governance and the creation in 2001
of a think tank, the 'Centre for Good Governance'.[128]
It remains difficult, however, to assess what role (if any) DFID
played in the success of AP's reforms.
76. Vision 2020 has been a somewhat controversial
policy due to its strongly modernist approach. Its predictions
that within 20 years, the proportion of AP's population engaged
in farming would decrease from 70% to 40%, and that large-scale
farming using modern technologies would become the norm, have
been criticised for neglecting the needs of the existing rural
population.[129] DFID
has sought to counter criticism of its involvement with Vision
2020 by supporting livelihoods projects and poverty reduction
programmes as well as budgetary, governance and sector reforms.
It has worked to maximise the pro-poor impacts of the GoAP's reform
strategy. In light of AP's successful reforms, DFID is now planning
to taper its programme in the State. Nonetheless, we think that
questions remain about DFID's original decision to invest so heavily
in a single state. It is not clear why the focus state with the
lowest proportion of income-poor people has received the majority
of DFID's development assistance. The Department's allocation
of assistance seems even more extraordinary if it is considered
that AP has, for some time, been the focus of considerable attention
from other donors. In 2003/4 for example, loans and grants to
AP totalled £418m.[130]
Given the substantial aid investments which DFID has made in AP,
we are concerned by the paucity of evidence that these have delivered
pro-poor results.
77. DFID justify their policy of targeting reforming
states, such as AP, on the grounds that this ensures that DFID's
aid is "well used"[131]
and provides an incentive to other states to engage in reform.
Aruna Bagchee, a Senior Governance Adviser for DFID in India told
us, for example "there is the question of, do you work only
in those states which are difficult and where the money may not
be as effective versus working with those states where there is
reform and you want to commend further reforms."[132]
Other commentators have suggested that 'cherry picking' of states
perceived to have a commitment to economic and political reform
has been a risk management strategy on the part of donors who
are being forced to become more and more accountable to home constituencies.[133]
Providing support to states with pro-reform, market-oriented administrations
also entails the risk of DFID being accused of ideological bias.
We understand that it is easier for DFID to work in states characterised
by good governance. However, we think that DFID's decision to
direct a very large proportion of its India budget into a single
state programme ought to have been better justified, given the
considerable cost to the UK taxpayer involved. Also, as we have
observed in reference to DFID's broader aid allocation policy
elsewhere, targeting the best performing states leaves the question
"what is on offer for the poor performers?"[134]
We are concerned that DFID's policy of supporting reforming states
should not exacerbate the already growing inequalities between
states in India.
78. DFID asserts that its projects and programmes
in India "are not considered significantly vulnerable to
changes in governments."[135]
During our inquiry we heard of concern that DFID's policy of supporting
states with reforming governments could be derailed by changes
of administration. These concerns were prompted by the change
of government which occurred following the 2004 state elections
in AP. The new GoAP has reduced the State's focus on the Vision
2020 strategy, and reversed some of the previous government's
policies which had been supported by DFID, most notably those
on power sector reforms (see paragraphs 58-60). Most policy areas
however, have not seen dramatic policy changes. Nonetheless, the
change of government in AP does highlight the risk which DFID
took in selecting its focus states on the basis of their pro-reform
administrations.[136]
79. Some witnesses told us that there was a question
over whether DFID was engaging with the right four states.[137]
Actionaid for example, told us that donors' state-focused support
in India had been provided on the assumption that different donors
would work in different states, but argued that this assumption
had not been born out. Furthermore they suggested that the withdrawal
of government-to-government financial assistance by all but five
other donors — in response to the NDA government's shift
in aid policy — should provide the impetus for DFID to reconsider
which states it should focus upon.[138]
We understand from DFID that they have no plans to develop any
new focus state programmes.[139]
They told us that this was because of the time it would take for
them to develop a partnership with a new focus state, and the
likely time limit on their future engagement with India. We also
heard that the GoI was becoming increasingly unhappy with donors'
growing desire to 'cherry pick' the most reforming states in which
to deliver their programmes.[140]
We agree that it would not be sensible for DFID to develop new
focus states at this stage, if it anticipates a complete withdrawal
from the country within the next ten years. If this is not the
case however, then we think that timescale should not be seen
as an impediment to the development of further focus state programmes.
Whether focus state programmes are the most effective way to deliver
DFID's assistance to other states is of course another issue.
80. A number of people we met during our visit identified
UP and Bihar as states in which they felt DFID should be engaging.
DFID staff themselves told us that India's MDG targets could not
be met if they were not met in UP and Bihar. Nearly half of India's
poor people live in these large and populous states, where service
delivery is dysfunctional, governance is poor and allegations
of criminality in politics are rife. Given the 'difficult environment'
which exists in these states, DFID has indicated it is unlikely
that it would wish to develop a 'partner' relationship with their
governments. Nonetheless it seems clear that in the future, effective
poverty reduction work in India will necessitate engagement with
UP and Bihar.
81. Any move by DFID to engage in UP and Bihar would
obviously require careful analysis and planning. It might draw
lessons from the World Bank's Lower Income Countries Under Stress
(LICUS) methodology, as well as from DFID's own work on development
interventions in conflict areas.[141]
We were pleased to hear that DFID has established a Working in
Difficult Environments (WIDE) task team to look into the possibility
of expanding its engagement with the two states. It is encouraging
that the World Bank is also looking into this possibility. We
were impressed by the polio eradication programme which DFID is
funding UNICEF to deliver in UP,[142]
and believe that working in this way through partner agencies
may be one of the most appropriate and effective ways for DFID
to expand its engagement with UP and Bihar. Engaging other organisations
as 'drivers of change' to work on DFID's behalf makes particular
sense where these agencies have already established a comparative
advantage by working in particular locations. Another possibility
might be for DFID to work with individual government organisations
and specific district administrations (as the Dutch have, in the
past, done in Kerala). We recommend that in UP and Bihar, DFID
should work through certain, carefully selected state organisations
and district administrations, as well as with international NGOs
and UN agencies. The recent announced move by DFID to provide
funding to the State AIDS Control Societies of UP and Bihar[143]
reflects the model we have in mind here.
82. Another sensible strategy would be for DFID to
develop its work with civil society organisations, work which
it has already begun through its Poorest Areas Civil Society (PACS)
programme. Of the 89 projects currently funded through the PACS
programme, 24 are in UP and 19 in Bihar. During our visit to India,
we were told of two constraints on DFID's work with civil society:
- first, GoI concerns about donor
engagement with civil society organisations make such work bureaucratic
and time-consuming;
- second, engagement with civil society is intensive
in terms of DFID's human and financial resources.
As discussed further in paragraphs 123-124, however,
we think neither of these constraints should be seen as sufficient
to prevent DFID expanding its PACS programme, which we strongly
encourage it to do.
83. During our visit we observed marked disparities
in levels of poverty within individual states. It was put to us
that DFID's work might impact on the lives of more poor people
if the Department were to target specific pockets of poverty in
a wider range of states than its current four. Dr. Michael Lipton
told us however, that "There would be a loss of focus if
DFID were to say they will go to any pocket of poverty. Supposing
that instead of four states you had 20 pockets of poverty, it
would be very difficult to get an effective management of a programme
like that."[144]
The Asian Development Bank (ADB), which has also taken a focal
state approach, endorsed this view.[145]
84. The ADB also commented however, that "while
the 'focal state approach' has several advantages, donors should
ensure that there is a fair spread of their assistance across
states."[146]
DFID told us that they endeavour to achieve such a 'fair spread'
by working with NGOs and through their National Programme. During
the course of our visit to India we were shown evidence of massive
poverty in states where DFID does not have focus state programmes.
We were therefore pleased to hear that under its 2004-2008 India
CAP, DFID is "refashioning a strengthened National Programme".[147]
We were surprised to learn, however, that DFID is planning to
almost double the budget of its National Programme in the next
financial year from £79 m. in 2003/4[148]
to £147 m. in 2004/5.[149]
Given what DFID told us about the success of its individual state
programmes, we feel that the Department has not adequately justified
and explained its decision to double its spending through its
National Programme in 2004/5. We hope that this funding decision
has been made on its own merits and not determined by the GoI's
changing priorities on the provision of aid.
Working with the World Bank and
other agencies
85. DFID has worked with a number of UN agencies
and bilateral donors, as well as with the Asian Development Bank
and World Bank. In each partnership they seek to take account
of the comparative advantages of their partner agencies. Following
the GoI's 2003 shift in aid policy (discussed in paragraphs 32-33),
many small bilateral donors are now in the process of withdrawing
from India. This has had the effect of diluting donor voice on
issues including gender and human rights. Of the six bilateral
and two multilateral donors which have been asked to remain, only
the World Bank's programme is similar to DFID's in terms of policies
and instruments (see Table 3). Charlotte Seymour-Smith told us
that DFID in India has "a very close relationship with the
World Bank and constant dialogue with them." She argued that
this collaboration strengthens the work of both organisations,
allowing DFID to "broaden and deepen [the Bank's] focus on
poverty reduction."[150]
It seems likely that the World Bank's recent decisions to move
away from a focus state approach and to investigate the possibility
of expanding its engagement in UP and Bihar,[151]
have influenced DFID's thinking on these issues, which, as noted
above, appears to be moving in the same direction. Table
3: Remaining donors
