Annual Review Project Title: KENYA SOCIAL PROTECTION PROGRAMME Date started: 2007 Date review undertaken: 05/2013

superfluitysmackoverΑσφάλεια

23 Φεβ 2014 (πριν από 3 χρόνια και 5 μήνες)

217 εμφανίσεις



Annual Review


Project Title:


KENYA
SOCIAL PROTECTION PROGRAMME


Date

started
:

2007



Date review undertaken:

05/2013




Instructions to help complete this template:


Before commencing the annual review you should have to hand:




the Business Case or ea
rlier project documentation.



the Logframe



the detailed guidance (How to Note)
-

Reviewing and Scoring Projects



the most recent annual review (where appropriate) and other related monitoring reports



key data from ARIES, including the risk rating



the separate

project scoring calculation sheet
(
pending access to ARIES
)


You should assess and rate the individual outputs using the following rating scale

and description
.
ARIES
and the separate project scoring calculation sheet will

calculate the overall output sco
re taking account of
the weightings
a
nd

individual outputs scores:




Descript
ion

Scale

Outputs substantially
exceed
ed

expectation

A++

Outputs moderately
exceed
ed

expectation

A+

Outputs
met

expectation

A

Outputs moderately
did
not meet

expectation

B

O
utputs substantially
did
not meet

expectation

C



Glossary of Terms: Acronyms and Abbreviations

AR


-

annual review

BWC


-

beneficiary welfare c
ommittee

CBT


-

community based targeting

CT


-

cash transfer

CT
-
OVC

-

Cash Transfer

Programme for Orphans an
d Vulnerable Children

DFID


-

Department for International Development

DP


-

development partner

GoK


-

Government of Kenya

HH


-

household

HSNP


-

Hunger Safety Net Programme

M&E


-

monitoring and evaluation

MIS


-

management information system

NDMA


-

G
overnment of Kenya
National Drought Managem
ent
Authority

NGO


-

non
-
governmental organisation

NSNP


-

National Safety Net Program

NSPP


-

National Social Protection Policy

PMT


-

proxy means test

PSP


-

payment service provider

SLA


-

service level agree
ment

SP


-

social protection



1

Introduction and Context


What support
is

the UK provid
ing
?

1.
DFID is providing support to the
establish
ment of

a government
-
led national system for long
-
term and
guaranteed
cash transfers (CT
s
)

to the poorest and most vulne
rable 10% of
households (HH
s
)

in Kenya.

Through a ten year £122m programme, approved in 2007,
DFID will support:



摥v敬潰m敮t of
愠獯捩慬 灲潴散ti潮
灯li捹 fram敷潲k

慮d im灬敭敮t慴楯n m散e慮i獭s
;



eff散tiv攠摥eiv敲y 潦ot桥 G潶敲湭敮t
-
l敤 programme of
C慳a
呲慮Tf敲
s f潲oOr灨慮s 慮d V畬湥ra扬攠
C桩l摲敮d(
CT
-
OVC
)
; and



灲潶i獩潮 潦o

猠t漠
桯us敨潬d
猠i渠獯ne of t桥 灯潲敳o 慲敡a wit桩渠n桥 ari搠慮搠獥mi
-
ari搠污湤猠of
K敮y愠a桲潵g栠t桥 H畮来r Safety N整 Programme

(
HSNP
)
.

2.
In 2012/13
,

DFID provided financial supp
ort and tech
nical assistance totalling
£21.4
m

against these three
outputs.

What are the expected results?

By 2017
:



a

獵獴慩湡扬攬 捯m灲敨p湳nv攠湡ei潮慬 獯捩慬 灲潴散瑩e渠獹獴em will 扥 i渠灬慣a
;



a

G潶敲湭敮t
-
l敤 湡ti潮慬
捡獨 tr慮sfer

獹獴敭ewill 灲pvi
摥d

猠t漠
㔰〬〰〠of t桥 灯潲o獴

猠i渠
K敮ya
;



f
i湡湣nng

will 扥 m潲攠 獵獴慩湡扬e

-

G潶敲eme湴n of K敮y愠 will 灲潶pd攠 8
5
┠ of t潴慬 f畮摩ng of


灲pgramm敳, 捯浰cr敤 t漠㈴┠i渠㈰n
5




m
潲攠famili敳ewill 扥 慢l攠t漠
獥s搠捨cl摲敮dt漠s捨c潬
-

t桥 灥r捥ctag攠of
捨cl摲敮d
i渠扥湥fi捩慲y


敮r潬l敤 i渠獣桯潬 will
i湣r敡獥st漠㤴┠
(
fr潭 㠷┠i渠㈰〷


慮a




b
敮efi捩慲a
桯畳u桯ld
猠li癩湧
i渠
t桥
湯rt桥r渠arid

慮搠獥浩
-
慲楤

捯cnti敳eof K敮y愠will
r整慩渠a湤/or
慣a畩re liv敳瑯捫
,

慮d
扥 l敳猠汩k敬y to f慬l
wit桩渠n桥 灯o
r敳t ㄰┠
of


i渠䭥nya
.


Wh
at is the context in which

UK support

is provided
?

3.
Poverty rema
ins widespread: according to the latest data, almost
46
% of the population
of Kenya
are
living below the poverty line
i

and 35% of children are chronically maln
ourished
ii
. There are marked
geographical variations, with some of the poorest areas being the arid and semi
-
arid lands in the north of
the country.

4.
The Government of Kenya

(GoK)
is committed to the development of a national social protection system

an
d this commitment

is
underpinned
by Kenya
’s long
-
t敲m 灬慮
-

si獩o渠㈰㌰
-

慮搠

t桥 C潮獴it畴楯n
E㈰㄰)K
e潷ev敲e c
潶敲eg攠
of 獯捩慬 灲潴散ti潮
r敭慩湳nl潷,

s
捨cm敳e慲攠fragm敮t敤, t慲geti湧 i猠灡t捨c
慮a

f畮摩湧 i猠i湳n捵牥
iii
.
DFID is one of several key
development partners in the social protection sector
:
others include

the World Bank, UNICEF and AusAi
d. The World Bank has played a particularly

important

role in
the past year in
facilitating the design of a consol
idated
CT

programme,
the National Safety

Net
Program

(NSNP)
,

that
focuses on five
cash transfer

programmes:
the
Cash Transfer

Programme for
Orphans and Vulnerable C
hildren (CT
-
OVC),
the Hunger Safety Net P
rogramme (HSNP),
the O
lder
Persons
Cash Transfer

(OPCT), the
Cash Transfer

for Persons with

Severe D
isabilities (CT
-
PwSD)

and the
Urban Food S
ubsidy
P
rogramme (UFS).

5.
Although national elections in March 2013 passed off

largely

peacefully, there was, nonetheless, some
disruption to programme delivery. For example, discussions about Government

budgetary allocations had
to await the new Government; and there were some delays to
HH

registration for phase two of the HSNP
programme

and to CT
-
OVC biometric enrolment
, as a result of simultaneous election registration.

6.
During the past year t
here h
ave
been important
changes in institutional arrangements for social
protection

and there are some continuing institutional uncertainties
.
The
dissolution
of
the Ministry

of State
for Northern Kenya and Other Arid Lands in March 2013
,

which was

previously
overseeing the HSNP
programme
,

led to the reallocation of this programme to the National Drought Management Authority

(NDMA)

in the Ministry of Devolution and Planning
and a need to build relationships with new
GoK
counterparts.

The CT
-
OVC programme has r
e
mai ned i n the Gender Department

togethe
r wi th other CT
programmes
, but
has mov ed to

the Mi ni stry of Labour and Soci al Securi ty.
W
hi lst i t

i
s clear
that the
on
-
going
devolution process
will

have important implications for all sectors,
it is not yet clear
exactly what


2

devolution will mean for the
social protection
(SP)
sector
, given that

SP remains a national priority
.

Section A: Detailed Output Scoring

Output 1: A long
-
term comprehensive, sustainable & co
-
ordinated social protection system
established in
line with Vision 2030 Social Pillar III and the National Social Protection Policy (NSPP)

Output 1
s
core

and performance description
:
A
-

Output met expectations

Progress
against

expected
results
:


Baseline
:

No social protection policy or recognition

of the right to social protecti
on.

Milestone
:
Social protection

policy approved, sector review undertaken
and implementation mechan
isms

established
.

This milestone will be considered in two parts
.

i)
Social protection policy approved, sector review undertaken
:
met expectations

7.
The Social Protection Sector Review
,

conducted in
2011/12

(and published in June 2012)

informed
t
he
deve
lopment of the
National Social Protection Policy (NS
PP)
, which

was approved in May 2012
. The

policy
comprises three components:
social assistance (of which
CTs

are part), health

insurance and social
insurance
.
DFID
-
funded technical input to the social ins
urance component of the review and policy was
valued
highly
by partners.


ii)
Implementation mechanisms established
:

met expectations

8.
There has been excellent progress with developing an implementation mechanism to take forward the
CT

element of the pol
icy and
less progress with other components.
During the first part of the year
,
DFID was
co
-
chair of the SP Working Group under the Social Sector of Vision 2030
, but this group was rather dormant

and partners would have liked to see DFID
more active in thi
s
role. In the second half of the year
DFID has
been much more engaged

in policy
and joint programming dialogue
. In particular, DFID
has provided
highly
valued
support
to the establishment of sustainable financing mechanisms
for cash transfers
and
to the
design of the

NSNP
,
a key
mechanism
for implementation of
the
CTs

element of the
NSPP
.

Key achievements during the period

9.
DFID has:



h
elped
positively shape the disbursement linked indicators
of the NSNP
and their protocols, in particular
in relation to

HSNP
;



provided funding for high quality
technical assistance for the NSNP
,

including for the single registry,

which will be critical to the future management and tracking of
the
proposed scale
-
up and ha
rmonisation
of CT programmes;



i
nformed
joined up
NSN
P design through sharing programme innovations, in particular lessons from the
HSNP payments system and
the
rights and grievances
component
;



a
lo
ngside the World Bank,
provided active support to the
SP

Secretariat in critical discussions with
Treasury over
future
GoK
allocations to
CTs
;



t
hrough increased dialogue, positively
contrib
uted to the
GoK
decision to contribute

to HSNP for the first
time, with 100% of the Ending Drou
ght Emergency Medium Term Plan;



h
elped further diversify
SP
funding
by facilitating
the entry of AusAid int
o the sector and providing h
igh
quality technical and managerial services
to AusAid
in the context of a
delegated co
-
operative
agreement.

Caveats to these
important
achievements
:

o

To date, the approach to addressing gaps in the skills

needed to take forward the NSPP has been rather
ad hoc and heavily reliant on short term TA (national and international). Future support should give more
attention to sustainably building and maintaining national capacity.

o

Despite recent progress in co
-
or
dination and lesson learning between CT programmes at national level,
field visits suggest that these processes are still much weaker at operational levels (at least in sites
visited); more attention could usefully be given to sub
-
national levels.

o

There

has been less progress with developing implementation mechanisms for non
-
CT elements of the
NSPP
-

health insurance, social insurance and non
-
CT elements of social assistance. Whilst this is not
primarily attributable to DFID, a more active contribution
as co
-
chair of the SP working group early in the
year might perhaps have supported progress
iv

and there is scope to do more going forward.

o

Despite the increased engagement with Government on HSNP in the second half of the year, the


3

imperative to focus on fi
nancing issues meant that the space for programming and policy discussion was
still somewhat limited;
and
there is an opportunity to expand this in 2013/14.

Impact Weighting (%):

20%

Revised since last Annual Review?
N

Risk: High

Revised since last Annual

Review?
N

Output 2:
Long
-
term predictable
CT
s to poor OVCs

Output 2 score and performance description:
A
-

Output

met expectations
.
With one indicator scoring
A and the other B,
it is proposed
that greater
weight
should be given
to the indicator of im
pact

on
beneficiaries

(scored A)

than to the indicator of
receipt of
transfers (scored B).

Progress against expected results:

Indicator

Baseline

2012 Milestone

Progress

Number of DFID funded
OVC
HH
s that receive
predictable and timely
CT
s


Number of ben
eficiaries
by sex

8,500

HH
s


individuals:

female: 24,383

male: 18,117

33,750

HH
s


individuals:

female: 96,815

male: 71,935

B
-

moderately not met

34,824

DFID
-
funded
beneficiary
HH
s

female: 99,896

male: 74,224

But transfers not always
predictable and tim
ely.

Proportion of children in
beneficiary HHs aged 6
-
17 years enrolled in
school

87%

(2007)

90%

(3 percentage point
programme impact)

A

met

(see below for discussion
of data)

Despite some operational challenges, there is very good evidence from randomi
sed control trials of positive
impacts of this programme on beneficiary households.

Predictable and timely transfers

10.
The programme management information system indicates that
34,824

beneficiary
HH
s were registered
to receive transfers funded by DFID i
n 2012/13, which is
3% above the
milestone
.

11.
However, p
redictability and timeliness
of
CTs

is
important

in order

to max
imise the impact of a given
transfer value
v

and between October 2011 and September 2012
only 29% of CT
-
OVC transfers were paid on
time
vi
, mainly due to
delays in transfers from Treasury
.

Other delays were as a result of delays from
development partner’s transfers and operational difficulties within the Children’s Services Department. Under
the NSNP payments working group, these delays ar
e being analysed and systems to prevent them are being
established.


12.
Furthermore, during field visits, beneficiaries reported a substantial worsening of the predictability and
timeliness of payments since the launch of the
new
electronic payments pilot

system through Equity Bank in
October 2012. Cumulative delays from each payment cycle mean that the payment due in
January/February

was actually being made during field visits in
May
.

The
Equity pilot
is also facing challenges with
registration
and car
ding
of
beneficiaries.
8%
(1,849)
of those on
the previous
POSTA list
have
not yet
been biometrically
registered
for Equity payments; of those
registered
4%
(804)
hav
e

not yet received their cards, without whi
ch
they cannot access payments; and of those w
ith cards, 4% (706) did not access their payments in
February
vii
. 21% of
DFID
-
fun
ded beneficiary households (7,088) had
been issued with e
-
cards under the
Equity pilot

by March 2013
viii
.

13
.
Nonetheless, despite t
hese challenges
, beneficiaries met
during the AR

were unanimous in their
preference for the new payments system
through Equity
(as long as
problems
can

be resolved) due to
paypoints being closer and queues shorter.

It is also important to note that
both GoK and partners are aware
of the challenges and
are actively addressing them
; detailed analysis has been carried out and an action plan
was
drawn up as part of the Joint CT
-
OVC Review Mission.

14
.
District
Children's
Officers are
also well informed about the
issue
s with the pilot, though, interestingl
y,

mainly due to reports from Nairobi, rather than complaints from beneficiaries through beneficiary welfare
committees (BWCs)
,

as might be expected
.
This can be explained partly by
the
fact that
BWCs
are
fairly


4

newly established

and
not yet
fully
trained
; the BWC members met during the field visit
appeared to have a
rather weak grasp of their roles and of beneficiary rights.

It may also be linked to the finding from other
studies
ix

that
beneficiaries see CTs more as a gift than a right
, such that, accordi
ng to one of the

D
istri
ct
Children's Officer
s
x

met,

'beneficiaries do not generally

complain about not being paid'
.

Education impacts

15
. The evidence
on impact
is drawn from randomised control trials carried out in 2007, 2009 and
2011
xi
.


16.

Logframe
metric
:
The data suggest that the enrolment of children aged 6
-
17 years may have increased
slightly
(by
0.7 percentage points
)
between 2007
-
11 for beneficiary households, compared to controls.
However, this effect is neither as large as the target of 3 pe
rcentage points, nor statistically significant.

This
result is perhaps not surprising, given that

primary enrolment was already very high

in 2007

and
that there
are many reasons other than
cost

that constrain enrolment
xii
.

The reasons that this indicator i
s, nonetheless
rated as 'A' is that t
here
is

strong evidence of substantial positive impacts o
f the programme on

the enrolment
of sub
-
groups of children

for whom costs
are likely to
have been a key constraint
. For example, cash
transfers have resulted in:



a
n
increase of
19 percentage points

in primary enrolment amongst those
children
in beneficiary
hous
e
holds
living far (over 2km) from a primary schoo
l
; and



an
increase of
7.5 percentage points
in
secondary school enrolment

of children in beneficiary
househ
olds
xiii
.

17.
Furthermore there is evidence
of
other
very
positive
unanticipated

knock on
impacts

for

young people.



Young
people aged 11
-
16 at baseline (15
-
20 at follow
-
up)
are
7 percentage points less likely to have had
their sexual debut
;

and

young women
aged 12
-
24

are

5 percentage points

less likely to have ever been
pregnant than their coun
terparts in non
-
programme HHs
.
The impact on pregnancy is

driven primarily
by
an improvement

in
school
ing

attainment

of girls
.

xiv

18.
It should
also be noted that the i
mpact evaluations have found a wide range of positive impacts of the
CT
-
OVC programme beyond education, for example

between

2007
-
11 the programme resulted in
:



an i
ncrea
se

in total
household
consumption
,
of
Ks274
(approx £2)
per adult equivalent per month



i
mproved quality of food consumption
-

more eggs, fruit, meat and fish
-

for small and female
-
headed
households



improved health care and
increased measles vaccination
of
children aged 0
-
5
xv



substantial local economic spillover effects.
F
or every £1

transfer
red

local income increase
d

by
up to
£1
.81
xvi
.

19.
The nature of cash transfers programmes

means

that beneficiaries have considerable
choice
as to how
to spend their cash

and indeed this is a key advantage of such programmes. If beneficiaries

spend the cash
differently to the ways anticipated by outsiders, the impacts will also be different

to those anticipated
, but
may
,

as in this case
, be equally substantial
.

The redesigned logframe for phase two programming includes
three diverse indicators of impact on b
eneficiaries (instead of just one as here), which should enable
improved measurement of impact in future.

Targeting

20
. In order for achievement of this output to contribute to the programme purpose of an effective CT system
for the poorest / most vulnera
ble 10%
HH
s, it is clearly necessary that the programme targets these
HHs
.
Whilst

the
most recent
e
xternal
monitoring report
found that targeting is perceived as fair and transparent by
93% of beneficiary caregivers
xvii
, there are, nonetheless some issue
s

wi
th targeting.



Re
-
targeting/graduation
:
Some HHs joined the programme as early as 2007 and may have seen
substantial changes in their poverty status in the intervening period, but no re
-
targeting has yet been
carried out. During field visits it was observe
d that some HHs supported since 2007 had set up small
businesses, found jobs or now had older children able to offer support.



Geographical targeting
:
Operational constraints have made it difficult for the CT
-
OVC programme to
reach marginalised and remote c
ommunities, particularly in those districts without DP funding. For
example, in Turkana, lack of transport and payment points means that only communities along the main
road are supported.



Poverty targeting
: Preliminary findings from the

draft baseline re
port
(2013)
,

presented during the CT
-
OVC Joint Review Mission,
indicate
that only
30% of current beneficiaries
are extremely poor and 42%
are
poor
, suggesting substantial errors of inclusion of the non
-
poor.



5



Categorical criteria
:
The programme title sugg
ests that it supports
HH
s with orphans
and vulnerable
children
,
yet the way targeting criteria are drawn means th
at 96
% of beneficiaries are actually (single or
double) orphans. Orphans are only slightly more likely to be living in poverty than other chil
dren

and less
likely than children with disabilities
, yet other categories of vulnerable children are not included.
In
targeting new households, t
he targeting criteria might usefully be reviewed, to ensure that
more of
the
most vulnerable and poorest chil
dren

are reached.

Impact Weighting (%): 45%

Revised since last Annual Review?
N

Risk: High

Revised since last Annual Review?
N

Output 3:
CT
s to the poorest and most food insecure
HH
s in Arid and Semi
-
Arid land areas (ASALs)

Output 3 score and performan
ce description: A+ Output moderately exceeded expectations

Progress against expected results:

Indicator

Baseline

Milestone (2012)

Achievement

Number of DFID funded
HSNP
HH
s that receive
CT
s

Number of beneficiaries
by sex

36,000
HH
s(2009)


Individuals:

Female: 131, 932

Male: 127,268

59,000
HH
s

Individuals:

Female: 208,578

Male: 216,222


A+
+

substantially

exceeded

69,000 beneficiaries funded

Female: 243,9304

Male: 252,870


% of HSNP HHs in the
bottom decile nationally

(percentage

point
difference wi
th control)

n/a

6 percentage point
reduction

A+
moderately exceeded

10 percentage point reduction

% of HSNP HHs that own
any livestock
(percentage point
difference with control)

n/a

6 percentage point
increase

A
met

6 percentage point increase

21
. In a
ssessing performance of this programme it is important to take account of its experimental, pilot
nature and the challenging and remote locations in which it is working. Several stakeholders underlined that
a key achievement of HSNP is to have proved that

CTs can work even in difficult locations.

Impacts

22
. The draft
report on the impact
evaluation
undertaken between 20
09

and 2012

(OPM, 2013)
xviii

provides
strong evidence
that

(in the context of a severe drought in 2011)

HSNP

has: increased

HH consumption

and
reduced extreme poverty
;
increased food consumption; enabled

HHs
to spend more on heal
thcare
;

and
increased

savings and the uptake of loans. Regarding the impact on

extreme poverty
,
b
eneficiary HHs
who have received
CTs
for two years are
10 percentag
e points less likely to fall within the poorest decile in
Kenya

than
if they had not received
the transfers
xix
. During 2012, control HHs (i.e. those not receiving
tran
sfers) became worse off; whilst
beneficiary
HHs
maintained their consumption levels and
th
e poorest
beneficiary HHs
actually
st
arted to improve theirs
.
The poverty gap and severity of poverty, which measure
how far below the poverty line
HHs

fall, have also decreased for HSNP HHs compared to control HHs, each
by

7 percentage points. When HHs
are disaggregated by size these poverty impacts are significant for
smaller HHs,
where the per capita value of the transfer is greater,
but not for larger ones.

23
. In terms of
retention and accumulation of livestock assets
, the overall results are encoura
ging but not
fully conclusive. Testimony from beneficiaries collected as part of the qualitative evaluation strongly supports
the conclusion that the programme has at least promoted retention of livestock
,

by enabling HHs to avoid
distress sales in the fa
ce of drought. Whilst quantitative analysis shows an increase in the proportion of
beneficiary HHs owning sheep and goats compared to control HHs, the causal impact of the programme
cannot be established, since this result does not hold up under further s
tatistical analysis
xx
.

2
4
. Commentary from beneficiaries during field visits supported the evidence on poverty impacts and
highlighted additional benefits, for example, promotion of
social inclusion
: “
Cash transfer
s brought some


6

families who were not bein
g visited into society. They are visited now. The community is more human”
(beneficiary, Turkana).

2
5
. There is also evidence of positive
gender impacts

of this programme.



Women are the majority (70%) of primary recipients

(
the named person on the front
of the beneficiary
card
)
xxi
.



At follow up 1 (in 2009) in 63% of HHs it was a woman who decided how the
CT

would be spent,
alone

or
in consultation with others;

though
by 2012, this had fallen back
a little
to 59%, associated with a general
decline in the co
ntrol over CTs by primary recipients
xxii
.



There have even been positive impacts on women's control of other cash within the household.
In both
2009 and 2012, the programme was found to have had a
small but statisti
c
ally significant impact on the
proportion

of ma
in budget decision
-
makers
in beneficiary HHs

who we
re female
.
xxiii



Community l
evel rights committees are
impressively gender
-
balanced, including in their leadership
(chairs and secretaries).
xxiv


Targeting

2
6
. The geographical targeting of this programme i
s very strong: it works in counties in which the vast
majority of the population lives below the national poverty line (96% in Turkana county). This means that
inevitably the majority of HHs receiving CTs are poor. Furthermore, those selected to receive

transfers are
slightly poorer on average than their neighbours. During phase one, three different approaches to targeting
beneficiaries were tested and the one shown to be the most effective, community based targeting (CBT), was
retained for use in phase

two. This will be combined with another approach, proxy means testing (PMT)
,

that
simulations showed to potentially outperform community based targeting in identifying very poor
HH
s. This
combination offers the potential to combine the best feature of b
oth methods, though there are risks (see
recommendations below)
.

Operations and Learning

2
7
. The programme has met a number of operational challenges:



All stakeholders agree that communications and co
-
ordination have been a key challenge over the past
ye
ar.



Lack of clearly defined responsibilities and shared operating systems mean that case management has
sometimes fallen between the cracks. For example, one community visited had not received payments
since Decem
ber;
different implementing partners g
ave

contradictory explanations and the
F
inancial
S
ector
D
eepening Trust
, the organisation contracted to oversee payments, only b
ecame aware of this
particular
issue as a result of the annual review field visit.



It has not been possible to track the overall
t
imeliness and predictability of payments

and for this reason it
was not possible to include timeliness/predictability
within the indicator for the number of beneficiary HHs
receiving CTs

(unlike for output 2).



Registration of HHs for phase two of the prog
ramme was supposed to have been completed by March
2013, but is running behind schedule due to problems with software design (compounded by rains and
elections).

2
8
. Most of these operational challenges are well understood by the team and stakeholders an
d remedial
measures are incorporated in the design of phase two. For example
:

indicators on timeliness and
predictability of payments are included within the logframe of HSNP and will be actively monitored;

the
contract for the payment service provider (P
SP) will have clear standards and penalties for non
-
compliance
set out in a service level agreement (SLA)
;
and case management will be underpinned by an improved
management information system (MIS), accessible by all stakeholders.

Impact Weighting (%): 3
5%

Revised since last Annual Review?
N

Risk: Medium

Revised since last Annual Review?
N

Section B: Results and Value for Money.

1. Progress and results

1.1 Has the logframe been updated since last review? Y

2
9
. The logframe was updated at a late stag
e once it became apparent that the latest version did not


7

incorporate changes recommended in the 2011/12 annual review. Such late revisions are clearly not in line
with best practice and any recommendations of the current review
should

be incorporated imm
ediately, in
order to avoid similar problems in future. Strenuous efforts were made to ensure that the changes reflected
as far as possible the targets envisaged earlier in the year, including by referencing documents prepared in
Oct 2012. Having carried

out a thorough assessment of all key achievements and weaknesses of the
programme (beyond the specific indicators in the logframe) the independent consultant
who carried out the
review was

confident that the ratings
wer
e a fair assessment of progress.

1.2

Overall Output Score and Description:

A
-

Outputs met expectations

30
. This is arrived at by averaging output scores. Outputs 1 and 2 met expectations. Output 3 exceeded
expectations.

1.3 Direct feedback from beneficiaries

31
. Meetings were held wit
h beneficiaries in four communities: one urban, one peri
-
urban, one rural close
to town and one remote rural. The key issues emerging from these discussions were the following:

i)
Beneficiaries strongly prefer cash to food assistance.


'Cash enables
HH
s

to make a choice'

(female beneficiary Turkana).

'
Cash transfer
s have helped because we were overwhelmed by our responsibilities as men, because
the droughts had decimated our herds and we have many wives. The money is not enough to
respond to all the m
any needs, but we are better off than others
,
' (male beneficiary, Turkana).

ii)
Household
s use cash in a wide variety of ways depending on their priorities
. The most frequently
reported uses in communities visited were for school uniforms
,

school books, s
econdary school
fees/transport, healthcare and food
xxv
.

In the urban area beneficiaries also use the transfers to buy water,
clothing and shoes; and in the remote rural area they buy tarpaulin to rainproof houses. Some invest in
small bus
inesses, pay

off de
bts or save
.

Two benef i ci ari es

menti oned usi ng CTs
to buy tradi ti onal beads
for thei r

daughters, highl i ghti ng a possi bl e negati ve i mpact of transfers
,

gi ven the associ ati on
of dowry
beads
wi th earl y marri age and sexual debut, though there i s no evi dence t
hat such use

of CTs

is
wi despread.

i i i )
Beneficiaries would like to see some design improvements
, i ncl udi ng:



cl oser paypoi nts (f or remote communi ti es)
;



more ti mel y and more frequent payments
;



faster resol uti on of compl ai nts and probl ems (such as stol en/mal
f uncti oni ng cards and outstandi ng back
payments)
-

several cases were reported as outstandi ng f or months or even years
;



trai ni ng for the rights commi ttees on the Kenyan Consti tuti on and ID cards f or the commi ttee members,
so that thei r rol e i s f ormal l y rec
ogni sed by the communi ty and vi si tors
; and



compl ementary servi ces, so that CTs do not get absorbed i n payi ng f or emergency heal thcare
;


(Thi s comment i nteresti ngl y f l ags the potenti al synergi es between heal th i nsurance and
CT
s. Health
insurance for the po
orest would limit lumpy health expenditu
re and increase

the possibi
li
ty of CTs being
invested in productive activities
.)


1.4 Summary of overall progress

Output 1
:

32
. The
social protection review and policy

have been finalised as planned.
Excellent pr
ogress has been
made in developing an implementation mechanism for the CTs component of the policy
-

through the design
of the NSNP
-

though less progress is noted in other elements (e.g. health insurance). DFID's engagement
in SP policy processes has bee
n much stronger in the second half of the year than the first; and in recent
months DFID has provided
very
active and valued support to the
establishment of sustainable financing
mechanisms
for cash transfers
and to the
development of the NSNP.

Output 2
:

33
. The
CT
-
OVC programme

has provided CTs to the intended number of
HH
s, though these transfers
have not always been timely or predictable. There is strong evidence of positive impacts on education,
predominantly at secondary level, with unanticipated kno
ck
-
on impacts in reducing risky sexual behaviour of
adolescents and pregnancy of girls. There are some issues with targeting and timeliness/predictability of


8

transfers that need to be addressed in order for achievement of this output to fully contribute t
o the
programme purpose of establishing a CT system for the
poorest and most vulnerable 10%
.


Output 3
:

3
4
. The innovative
HSNP

has proved that CTs can work even in challenging and remote parts of Kenya.
Lessons learned from the pilot phase will inform t
he design of this and other CTs going forward.
During a
period in which
HHs

not receiving transfers (the controls) have

become worse off
, t
he programme has
protected beneficiary
HH
s, enabling them to
retain their consumpt
ion levels and livestock assets.
The
poorest
beneficiary households have actually increased their consumption as a result of the transfers.
These impacts are seen especially for smaller
HH
s.

1.5

Key Challenges

3
5
. Key challenges with systems, co
-
ordination and capacities have been iden
tified and analysed; they have
informed the design of phase two DFID programmes and of the NSNP. For example:

Systems:

-

MISs
: As noted above
,

the lack of integrated management information systems (MISs) has
impeded
effective management of complaints and
grievances in both HSNP and CT
-
OVC and will be addressed as
part of HSNP

2

and SPP2
.

-

Payments
: Timeliness and predictability of payments has clearly been an issue for CT
-
OVC as detailed
above. It has not been possible to measure timeliness for HSNP in

phase one, a
key
gap that will be
addressed going forward.

-

Targeting
: Opportunities to improve the cost
-
effectiveness of both CT
-
OVC and HSNP programmes
through improved targeting have been identified.

Co
-
ordination

within and
b
etween CT programmes


-

Co
-
ordination between HSNP implementing partners has been a challenge, as has co
-
ordination between
different CT programmes at both national and sub
-
national levels. Co
-
ordination
/

harmonisation is a key
focus of the NSNP.

Capacity

-

Decentralised level
s
: Limited operational budgets and uneven allocation of these between DP
-
supported
and other areas has constrained CT
-
OVC a
bility to operate effectively in some areas

and
this
is now being
addressed
.

-

National
: Capacity in the SP and CT
-
OVC secretariats i
s also seen by staff there as a key constraint

and
will be addressed as part of the NSNP.

-

HSNP
: A key challenge going forward will be promoting and capacitating GoK ownership of HSNP. This
may be exacerbated by the level of ambition of some aspects of p
hase two of the programme (
see
recommendation 16
)
.

1.6 Annual Outcome Assessment

3
6
. Huge strides have been made in the past year towards the project purpose of establishing a
government
-
led national system for long
-
term and guaranteed CTs to the poorest

and most vulnerable 10%
of
HH
s in Kenya. The most important developments are:



Parliamentary approval of the NSPP (2012)
, which

articulates the vision to provide universal access to
social protection for the vulnerable throughout their lifecycle. GoK will

work progressively to meet these
rights and
to
establish institutional arrangements for

policy

implementation.



The design of a NSNP (2013
-
17). Development partners will provide coordinated support to enable the
GoK to: increasingly harmonise the five
CT

programmes currently under operation; strengthen systems
for targeting, registration, payments and monitoring; and expand their coverage. The design is
underpinned by high quality technical analysis and extensive consultations and as a result will address

most of
the
challenges and weaknesses with current CT programmes.

3
7
. These developments have transformed the prospects of achieving the programme purpose by 2017,
which is now very likely indeed, notwithstanding the challenges flagged in previous section
s.

2. Costs and timescale

2.1 Is the project on
-
track against financial forecasts: Y

3
8
. The programme spent £21.4m in 2012/13 against a budget of £18.3m. The overspend was because in


9

2011/12, £3m was withheld due a fiduciary risk assessment issue. T
he issue was resolved and the funds
were released in 2012/13.
2.2 Key cost drivers

3
9
. Transfers account for the majority of the budget, so key cost drivers are the number of beneficiary HHs
and the size of transfers

(see paras
52
-
53
)
.

2.3 Is the projec
t on
-
track against original timescale: Y

40
. In terms of the increase in the number of HHs reached and the impacts achieved for beneficiaries, the

project is on

track against the timescale set out in the original logframe. Some operational elements are
d
elayed. A key activity under output 3 in 2012/13 has been the registration of all HHs in the four counties of
northern Kenya in preparation for phase two of the programme. This was due to be finished by March 2
013,
but was only 40% completed
at the time
of the AR visit in April and
was completed

by the end of
July
2013
.

Due to effective mitigating action, this delay is not expected to result in the project being off
-
track against its
timelines in future.

3. Evidence and Evaluation

3.1 Assess any cha
nges in evidence and implications for the project

41
. High quality impact evaluations are rapidly expanding the evidence base on CTs, both in Kenya and
globally. There is now a strong body of global evidence that CTs r
aise living standards by

reducing pov
erty,
hunger and inequality; and that they help HHs sustain and improve livelihoods in the face of vulnerability
and shocks. There is evidence of their positive impacts on food consumption, education and healthcare
xxvi
.
Randomised control trials carried out

in Kenya between 2007 and 2011 demonstrate these impacts in the
Kenyan country context
xxvii
.

42
. Given this, the DFID Kenya programme might consider focusing future evaluations on areas in which
the country and global evidence base is still weaker and where e
vidence could inform critical strategic
decisions in Kenya. Taking account of issues highlighted in DFID guidance to country offices,
Evaluating
Social Transfer Programmes
, 2012, and of the findings of this annual review, such areas might include:



whether

and how CTs impact women's control of cash within the
HH

and women's wider empowerment;



what combination of interventions (including CTs) is most effective in addressing high chronic child
malnutrition;



the cost
-
effectiveness, and relative impact of diffe
rent transfer levels; and



the economic spillover effects of CTs in the northern counties and the optimal package of
complementary activities to support livelihoods.

43
. To complement on
-
going quantitative evaluations, theory or case based approaches might
usefully be
employed to explore how and why effects come about.

3.2 Where an evaluation is planned what progress has been made?

4
4
. Randomised control trial evaluations have been implemented from the outset of both the programme of
CT
-
OVC and the HSNP and
their findings are referenced throughout this review. Recent reports are listed
here
xxviii
.

4. Risk

4.1 Output Risk Rating: Low/Medium/High

4
5
.
Two programmes have been developed to take forward the outputs of SPP1. SPP2 covers outputs 1
and 2 and HSNP
2 addresses output 3.
The forward looking risk rating
for both programmes is

Medium
.

4.2 Assessment of the risk level

4
6
. The key risks
(high likelihood, high impact) identified in the risk matri
ces for phase two programming
include:

-

risks of Government
being unable to deliver on policy or financial commitments
;

-

delays caused by post
-
election institutional change at national and decentralised levels
; and

-

risk of CT
-
OVC funds being diverted away from intended beneficiaries, due to weaknesses in the
payment
s system (see section 4.3).

4
7
.
These risks appear to be correctly identified and rated. Appropriate m
itigating measures are in place to
address
them
and the res
idual risk rating
of
medium

seems correct.

The reduction in risk rating from high to
medium f
or
output 2 (the
CT
-
OVC programme
)

is explained by the substantial efforts to address key risks,
including
the introduction of electronic payments to
beneficiaries to
reduce the risk of monies going astray.



10

4
8
.
DFID has also identified as a
key

risk
for HS
NP2
that
those
phase one
HSNP
beneficiari
es
no longer
eligible in phase two
,

who are
exited
from the programme
,

feel

disgruntled.

F
orthcoming decisions will
impact this risk

and t
his
AR
makes two recommendations pertinent to
managing

it
:

-

ensure that t
he detailed design of the new targeting methodology (which will combine
PMT and CBT
data)
is
transparent and
comprehensible to communities

(recommendation 12)
;
and

-

review
the level of ambition
of the rights component
and
ensure that
staffing
is adequate

for complaints to
be addressed quickly and effectively

(recommendation 16)
.

4.3


Risk of funds not being used as intended

4
9
.
CT
-
OVC
: An in
-
depth financial audit covering the period Jan 2008


Jun 2011 was carried out and
finalised at the end of 2011.
There

was no evidence of fraud within the programme. The shift to electronic
payments with two factor
authentication

(the Equity
pilot
)

is aimed at reducing

the
risk
of transfers being
diverted away from the enrolled beneficiaries.


Regular financial reviews a
re also carried out by the World
Bank
; and
t
he Joint Review Mission (June 2013) concluded that there are no outstanding audit issues and
that
the
CT
-
OVC is
in full compliance with
audit reporting requirements.


50
.

HSNP
: A fiduciary risk assessment (FRA) w
as carried out in 2010
and
found that DFID funds were not
exposed to significant risk.
No funding is going through a third party provider or government systems. DFID
retains direct control of monitor
ing and measuring the funds flow;
CTs are transferred ele
ctronically
;

and
beneficiaries use a two factor authentication process to access the funds.

A further
FRA
on the payments
process
will be conducted
by FSD in

August

2013

and an audit of HSNP Phase 1 will take place in
October
/November

2013.

4.4

Climate
and Environment Risk

51
.
CT

programmes

have the potential to reduce vulnerability and increase resilience to climate variability
and change. There is strong evidence that the HSNP

programme

has played a key role in protecting
beneficiary
HH
s from the rece
nt major drought (see output 3 above); and the design of a system to scale
-
up
the transfers in response to shocks in phase 2 can be expected to further strengthen this protective
mechanism.

5
. Value for Money (VfM)

5.1 Performance on VfM measures

Cost
-
efficiency

(administrative costs as a proportion of total programme costs)

52
.

CT
-
OVC:

Total costs comprise
:

i)
the cost o
f CTs received by beneficiaries;

and ii) administrative costs
required to target and deliver them
, including

set
-
up, roll
-
out, oper
ations

costs and M&E.
The share of
operational costs in total costs has declined over recent years as set
-
up costs have diminished and the
programme has scaled up; by 2011/12
xxix

operational costs were only 11% of total costs
xxx

(from all sources)

which is
well

within the average range of
well
-
executed
CT programmes in developing countries
xxxi

.

This is
predicted to rise
by 5 % points
to an average of 16% over the four years of phase 2 (2013
-
17)

due to

important

investments in improving programme effectiveness
. Wh
ilst measured cost
-
efficiency will, thereby
fall slightly, cost effectiveness is expected to increase due to the anticipated benefits of these investments
(as outlined
in para 54
).


53
.

HSNP:

A
dministrative
costs are anticipated to
be on

average 21%

during phase 2 (2013
-
17)
.

xxxii


A
s a
result of the up
-
front costs of registration, targeting and improved payments systems

for phase two, admin
costs will be higher in the first year.
To justify
the

high costs of these investments
it will be important to
ens
ure that their potential benefits are fully realised (see recommendation 12

on targeting and
recommendation
15 on payments
).

Cost
-
effectiveness

(costs of reducing poverty through transfers)

5
4
.

CT
-
OVC:

Despite impressive performance
to date
on cost
-
effici
ency

measures, DFID
value
-
for
-
money
(Vf
M
)

analysis suggests there
to be scope to improve the cost
-
effectiveness

of this programme (poverty
impact for every £ spent)
.

For example, improving targeting would increase the proportion of CTs reaching
the poorest
; and s
trengthening the payments system should
i
mprove

the timeliness and predictability of
transfers
, thereby reducing

private costs to beneficiaries of repeated visits to collection points and

increasing

the consumption smoothing impact of tran
sfers.
Pr
ovision of high quality TA, as envisaged under
the
N
SNP, will support these improvements
.




11

5
5
.

HSNP
:

A s i mpl e pr oxy meas ur e of c ost ef f ec t i venes s i n
r educ i ng pover t y
was devel oped by DFI D
Kenya
x x x i i i

and o
n t hi s meas ur e
,

i n 2012/13
,

a s pend of £2.94 was r equi r
ed f or a £1 r educ t i on i n pov er t y
x x x i v
.

This relatively high cost is largely due to
important
one
-
off programme investments in registration, targeting
and the payments system

during 2012/13
, which
should
contribute to sustained improvements in cost
-
effectivene
ss
.
Indeed c
ost
-
effectiveness
is
anticipated
to
improve

in future years
(to
£1.30 per £1 re
ducti
on
in poverty b
y 2014/15), driven largely by

a reduction in
administrative costs (as one
-
off investments come to
an end)

and an increase in
the proportion of t
ransfers reaching the poorest
, as a result of improvements in
targeting
.

5
6
.
Every £ spent by DFID on strengthening programme systems will benefit
all

beneficiary households, not
just DFID
-
funded ones; so p
lanned increases in the number of beneficiary h
ouseholds receiving
GoK
-
funded
CTs through both programmes should further increase the
poverty impact of a given DFID spend
and hence cost
-
effectiveness.

5.2
Commercial Improvement and Value for Money

5
7
.
DFID funding to the CT
-
OVC programme is passed th
rough a World Bank Trust Fund. The overhead
charge is only 1%, representing good
VfM
.

5
8
.
The contract for the
payment service provider (
PSP
)

of HSNP has been put out to tender in order to
ensure a commercially competitive solution. Lessons learned in

phase one mean that the contract in phase
two will
have a

service level agreement

(
SLA
)
and that financial penalties will be imposed on the contractor
for non
-
compliance. During 2012/13 the CT
-
OVC programme has also started to use a commercial
PSP
.

5.3
R
ole of project partners

5
9
.
GoK
is the key partner on
SP

policy development and implementation and on the design and delivery of
the CT
-
OVC programme.

Government
-
wide public financial management reforms and work by the SP
sector over the past year have re
sulted in improvements that will facilitate improved VfM analysis in future
years.


60
.
DFID directly procures all services for the HSNP programme. M&E, payments and management/co
-
ordination were initially contracted through separate competitive tenders,

though by 2012/13 only the M&E
contract
(with O
xford
P
olicy
M
anagement
)
was on
-
going. Accountable grants were provided to NGOs for
two additional components: registration
(OXFAM
-
led consortium)
and social accountability

(HelpAge)
.

5.4 Does the project

still represent Value for Money : Y

5.5 If not, what action will you take?

Not
applicable

6. Conditionality

6.1
Update on

specific conditions

61
.

Not
applicable

7
. Conclusions and actions

62
.
The
SPP phase one
has continued to progress well. R
emaining challenges are well understood and
lessons have been learned. The next phase

incorporates substantial design improvements to address the
majority of issues identified in phase one.
Thes
e are
described in two
approved

business cases
, which
separa
tely cover OVC and HSNP, and which are available on the DFID website.

The recommendations
listed below

highlight those few
issues that are no
t
fully
addressed
in

current proposals.
These
recommendations are more fully explained and discussed in the accomp
anying report
Recommendations
from the Annual Review of the DFID Kenya Social Protection Programme
.

Recommendations
:

Output 1: Social Protection System

(NB. this section incorporates all recommendations relevant across
programmes and, therefore, to the NSN
P)

1.

Contribute to development of CT scale
-
up plan, promoting alignment of resource allocation with
geographical distribution of poverty and vulnerability (immediate
,by all NSNP stakeholders
)

2.

In view of evaluation findings regarding more limited impacts on l
arger households, actively support the
development of feasible
options for
varying benefits by

household size (
medium
-
term
, by all NSNP
stakeholders
)

3.

Focus
future
evaluation on areas in which country and global evidence base is weaker and where


12

evidence co
uld inform critic
al strategic decisions in Kenya

(see section 3.1

above)

(medium
-
term

by all
NSNP stakeholders
)

4.

Promote policy linkages between CTs and other social policy interventions to maximise impacts for the
poorest/most vulnerable (medium term
, by D
FID, WB, UNICEF and SP Secretariat
)

5.

Develop a sustainable
,

national training/mentoring system for the social protection sector (medium term
,
by all NSNP stakeholders
)

6.

F
ocus more capacity building support at
sub
-
national
levels of Govern
ment and community c
ommittees
and p
romote lesson learning between programme
field
staff

(medium
-
term
, by SP Secretariat and
NDMA
)


7.

Consider the potential for using mobile technology
to communicate with beneficiary committees and
for
monitoring payments and
beneficiary
satisfa
ction (medium
-
term
, by all NSNP stakeholders
)

Output 2: CT
-
OVC:

8.

S
upport a Government
-
led review of the criteria used to define 'orphans and vulnerable children', in
order that more of the poorest and most vulnerable children are reached

(short
-
term
, OVC Se
cretariat,
SP Secretariat, WB, DFID, UNICEF
)

9.

Support CT
-
OVC to learn from HSNP on the registration of polygamous families (short
-
term
, OVC
Secretariat, NDMA, DFID
)

10.

Link CT
-
OVC beneficiaries to complementary services (medium term
, by OVC Secretariat, DFID,
WB
)

11.

Ensure that resources are committed for periodic re
-
targeting and support Government to develop a
strategy for graduation, which involves systematically linking exiting households to appropriate
complementary services. Promote clear and consistent com
munications around exit and graduation
(medium term
, OVC Secretariat, SP Secretariat, DFID, WB, UNICEF
).

Output 3: HSNP

12.

Ensure that the ways in which data
are

combined in the final
combined
targeting methodology is
effective both in targeting
the
poorest

H
Hs

and in underpinning the rights component of the programme
(see complementary report for detailed recommendations)

(immediate
, DFID, NDMA, HelpAge
)

13.

Analyse
how cash is controlled within the
HH
,
in particular
the extent of women's control and t
he factors
that influence this and use
findings to
inform
programme design.

( C o n s i d e r a c r o s s
-
p r o g r a m m e
a n a l y s i s,
a l s o c o v e r i n g C T
-
O V C a n d p o s s i b l y o t h e r C T s.)

( m e d i u m
-
t e r m
, D F I D, N D M A
)

14.

In line with
the
improved governance proposal for HSNP2, engage
GoK
increasingly

in programme
design decisions to promote ownership and susta
inability (medium
-
term
, DFID, NDMA
)

15.

Ensure that the minimum standards in the contract and SLA for the payment service provider are
sufficiently stringent and that penalties are enforceable.
(imme
diate
, FSD, Equity Bank
)

16.

R
eview the
level of ambition of the rights component, in view of both
field
-
level

capacities available to
deliver it during phase two and the commitment to progressive hand
-
over to GoK

(short
-
term
, DFID,
NDMA, HelpAge
)

8
.
Review
Process

63.
The review was carried out by an independent consultant, with the active input of the DFID Kenya
Poverty, Vulnerability and Hunger Team,
GoK
, other development partners
-

including AusAid, World Bank
and UNICEF
-

and HSNP implementing partners
. Field visits were carried out
during April/May
to HSNP
project sites
in Lodwar and
Nakitoekirin, in
Turkana

County

and
to
CT
-
OVC project sites in
Kawangware,
Nairobi and
Muthetheni,
Machakos
. During field visits
,

meetings were held
with beneficiaries,
County and
District
GoK
officials and implementing partners.
The consultant
returned in June to participate

in the CT
-
OVC Joint Review Mission
, in order to promote
a
harmonised
approach to programme
review
.

Two
presentations of draft findings were made to
stakeholders
and feedback was incorporated.





i

Kenya National Bureau of Statistics, 2007. Kenya Integrated Household Budget Survey 2005/06,

Basic Report on Well
-
being in
Kenya.

ii

Kenya National Bureau of Statistics,

2008
-
09.
Demographic and Health Survey.

iii

CT
-
OVC draft Business Case 2012



13






iv

GoK did not call any meetings of this group

and in the second half of the year the co
-
chair responsibility shifted from DFID to the
World Bank. At around the same time, t
he NSNP

design process

was launched
, led by the World Bank and GoK Social Protection
Secretariat
;

and the working groups around this became de
-
facto the donor group
.


v

See for example
An Impact Evaluation of Ethiopia's Productive Safety Net Programme,
Gilligan e
t al, 2009, for evidence in this.

vi

World Bank, 2013.
Kenya National Safety Net Programme Technical Assessment.

vii

From figures in presentation at CT
-
OVC Joint Review Mission
on 19/06/2013
by
Leonard Matheka, consultant. It should be
noted that problems w
ith
registration

are not entirely due to weaknesses in the new system, but sometimes indicate discrepancies
in the previous POSTA list, since where recipient details are out of date on the POSTA list, households cannot be registered.

There is not yet any
breakdown of the reasons for failure of carded households to access
payments
, which could include
beneficiaries being unaware of the payment period, or cards malfunctioning.

viii

Data from Agency for Development Research (ADR), the company contracted to c
arry out a review of the pilot,
March 2013
.

ix

Onyango
-
Ouma, W. and Samuels, F., 2012.
Transforming Cash Transfers:


Beneficiary and Community Perspectives of the
Cash Transfer for Orphans and Vulnerable Children Programme in Kenya
, ODI, University of Keny
a, DFID.

x

District

C
hildren's
O
fficer

in Muthetheni, Machakos

xi

Ward et al, OPM, 2010.
CT
-
OVC Operational and Impact Evaluation, 2007
-
09, Final Report
.


and
Presentation by University of North Carolina '
Impact of the Kenya CT
-
OVC on the transition to ad
ulthood (and other outcomes)'

xii

Analysis carried out by UNICEF suggests
that the principal reasons why children are out

of school vary substantially b
y

region
and include parental disapproval, labour obligations,
disability
/
illness, as well as costs
. Disa
bility
/
illness
is a

relatively important
factor

in districts where there is almost 100% attendance
.

xiii

Presentation by University of North Carolina

op cit.


xiv

ibid

xv

ibid (this and all impacts listed in bullets above)

xvi

Davis, FAO and University of North C
arolina. Presentation, April 2013, on '
Evaluating the Spillover Effects of Kenya’s CT
-
OVC
Program.

xvii

PCL and DIC Consultants
, 2013.
CT
-
OVC External Monitoring Report.

xviii

Hurrell and Sabates
-
Wheeler, OPM, 2013.
HSNP Quantitative Impact Evaluation Final Rep
ort (draft).

xix

This conclusion is reached through comparison of the change
between Nov 2011 and Nov 2012
for beneficiary households,
compared to the change for
the
control
group who
do not benefit from the programme.

xx

ibid

xxi

OPM, 2013, op cit.

xxii

ibid

xxiii

ib
id

xxiv

Women were 47% of members of rights committees, 1/3 of chairs and over 2/3 of secretaries
, according to HAI Turkana Office
records.

xxv

Food purchase is the most commonly reported

use of CTs in evaluations. Beneficiaries in the remote community 100km f
rom
Lodwar in Turkana mentioned food first in their list of uses of CTs, but
AR focus groups

in the other three communities visited did
not
mention food spontaneously, but only
confirmed
that they used CTs to pu
rchase it when directly asked. A possible ex
planation
for the lesser mention by the less remote communities is that these beneficiaries are somewhat less poor than average and ten
d to
spend
more of their CTs on non
-
food items.

xxvi

DFID, 2011.
Cash Transfer
s Literature Review
.

xxvii

Reports as follows:

-

Taylor et al, University of North Carolina, 2012.
Evaluating General Equilibrium Impacts of Kenya’s
CT

Program for Orphans and
Vulnerable Children.

-

Handa et al, University of North Carolina, 2012.
Impact of the Kenya CT
-
OVC on the Transition to Adulthoo
d.

-

Hurrell and Sabates
-
Wheeler, OPM, 2013.
Kenya Hunger safety net programme Monitoring and Evaluation Component,
Quantitative Impact Evaluation Final Report: 2009 to 2012 (draft)

xxviii

ibid

xxix

Data is not available on 2012/13 as the Kenyan fiscal year runs
to June

xxx

M. Mann, 2011.
Kenya OVC Cost Analysis Project

xxxi

O
perating costs of well
-
executed social transfer programmes tend to cluster in the range of 5
-
15%.
Hodges et al, 2011.
Guidance for DFID Country Offices on Measuring and Maximising Value for Money

in CT Programmes, p28.

xxxi

DFID Kenya Public Account Committee Return.

xxxii

DFID Kenya Public Accou
nt Committee Return.

xxxiii

On this measure t
he direct cost of transfers to the poorest HHs is assumed to result in a pound for pound reduction in poverty.
All other

programme spend, for example, on administrative costs and transfers made in error to less poor HHs is assumed to have
no poverty impact.

xxxiv

Implying that for every £2.94 spent, £1 go
es directly

to a poor, correctly targeted HH as a transfer. The other £
1.94 is split
between administrative costs and transfers that are mis
-
targeted to better off HHs.