October 2014
This publication was produced at the request of the United States Agency for International Development. It 
was prepared independently by Harry Blair, William Cartier, Akhter Hussain, and Naim Mostofa of Social 
Impact, Inc.
Performance Evaluation of Strengthening Democratic 
Local Governance (SDLG) Project
Cover Photo: Photograph of Ward Shava planning meeting courtesy of SDLG from RDRS. 
PERFORMANCE 
EVALUATION OF THE 
STRENGTHENING 
DEMOCRATIC LOCAL 
GOVERNANCE (SDLG) 
PROJECT 
MACRO-LEVEL ADVOCACY AND LOCAL-LEVEL PARTICIPATION 
DATE: October 2014
USAID Contract AID-OAA-I-10-00003
Task Order AID-388-TO-12-00001 
DISCLAIMER 
The authors’ views expressed in this publication do not necessarily reflect the views of the United States Agency 
for International Development or the United States Government. 
ACKNOWLEDGEMENTS 
The evaluation team was most fortunate in receiving all needed assistance from the Social 
Impact staff and the SDLG staff in Dhaka, and the partner NGOs who assisted with field 
logistics on its visits to Khulna and Rajshahi divisions. 
In particular, the team would like to thank Saiful Islam, Deputy Chief of Party at the Dhaka 
Social Impact office, and Md. Pervez, Operations Manager, for their support throughout the 
team’s work. At the SDLG office, the team wishes in particular to thank Jerome Sayre, Chief of 
Party, Zarina Rahman Khan, Deputy Chief of Party, and Abu Md. Mohsin, Team Leader for 
Citizen Participation. Finally, thanks also to Rumana Amin, the Contracting Officer’s 
Representative at USAID for the Bangladesh Democracy and Governance Programs Evaluation 
project, and Sherina Tabassum, the Contracting Officer’s Representative for the SDLG project 
itself. 
CONTENTS 
Contents ............................................................................................................................................................. v
Acronyms ........................................................................................................................................................... 1
Executive Summary .......................................................................................................................................... 3
1. Introduction ............................................................................................................................................... 8
2. The development Problem and USAID’s Response ......................................................................... 8
4. Evaluation methodOLOGY and Limitations .................................................................................... 12
5. Findings and Conclusions ..................................................................................................................... 15
6. Lessons Learned .................................................................................................................................... 35
Recommendations ......................................................................................................................................... 35
ANNEXES ....................................................................................................................................................... 39
Annex I: EVALUATION SCOPE OF WORK ........................................................................................ 40
Annex II: DOCUMENTS REVIEWED ....................................................................................................... 52
Annex III: EVALUATION INSTRUMENTS .............................................................................................. 55
Annex IV: INTERVIEW LIST ...................................................................................................................... 68
Annex V: Tables and Figures ....................................................................................................................... 74
Figure 1: SDLG Goals and Components .............................................................................................................................74
Table 1 Field Visit Plan ...............................................................................................................................................................74
Figure 2: SDLG Sample LGUs Visited ....................................................................................................................................75
Table 2: SDLG Evaluation Questions and Conclusions in Brief ......................................................................................75
Table 3: SDLG Spending on Support to LGAs March 2011 to December 2013 (in US$) ......................................77
Table 4: Comparison of the Percent of Union Parishad Respondents who Report Being Aware of UP Efforts 
to Raise Revenue (PMP 4.2) .....................................................................................................................................................77
Table 5: Comparison of the Percent of Paurashava Respondents who Report Being Aware of PS Efforts to 
Raise Revenue (PMP 4.2) ..........................................................................................................................................................77
Table 6: Comparison of the Percent of Union Parishad Respondents who Agree that the UP Manages their 
Funds with Transparency and Accountability (PMP 4.2) ..................................................................................................78
Table 7: Comparison of the Percent of Paurashava Respondents who Agree that the PS Manages its Funds 
with Transparency and Accountability (PMP 4.2)...............................................................................................................78
Table 8: Own Source Revenue Expenditure Patterns in 2012-13 .................................................................................78
Table 9: Comparison of Citizens’ Evaluations of the Quality of Road Construction, Repair, and Maintenance 
in Union Parishads (PMP 3.5) ...................................................................................................................................................79
Table 10: Comparison of Citizens’ Evaluations of the Quality of Maintaining Market Places for Citizens in 
Paurashavas (PMP 4.10) .............................................................................................................................................................79
Annex VI: Performance Management Plan for SDLG: Targets and Actual Results, 2011-2013 .. 80
SDLG Performance Indicators – Results of 2011 (Year 1), 2012 (Year 2) and 2013 (Year 3)........................ 80
Annex VII: Footnotes on Survey Based Indicator Values for 2012 (3.4, 3.5, 4.2, and 4.5) ........... 85
Annex VIII. Evauation Design Matrix ....................................................................................................... 88
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 1 
ACRONYMS 
ADB Asian Development Bank 
AL Awami League 
BDGPE Bangladesh Democracy and Governance Programs Evaluation (project) 
BUF Bangladesh Urban Forum 
BUPF Bangladesh Union Parishad Forum 
CiG Citizens in Governance 
COR Contracting Officer’s Representative
CTG Caretaker Government 
CUS Center for Urban Studies 
DG Democracy and Governance (sector of USAID) 
DLGP Democratic Local Governance Program 
EOP End of project 
GAF Governance Advocacy Forum 
GiZ Deutsche Gesellschaft für Internationale Zusammenarbeit 
GOB Government of Bangladesh 
ILLG Improving Local Level Governance (program) 
IQC Indefinite quantity contract 
JICA Japanese International Cooperation Agency 
LG Local government 
LGA Local government association 
LGD Local Government Department 
LGED Local Government Engineering Department 
LGSP Local Government Support Project 
LGU Local government unit 
LOE Level of effort 
LOP Life of project 
M&E Monitoring and Evaluation 
MAB Municipal Association of Bangladesh
MP Member of Parliament 
NCSLG National Constituency for Strong Local Government (program) 
NILG National Institute for Local Government 
OSR Own source revenue 
PMP Performance Management Plan 
PNGO Partner Non-Governmental Organization 
PPC Policy and Program Coordination Bureau 
PPL Policy, Planning and Learning Bureau 
PROGATI Promoting Governance, Accountability, Transparency and Integrity (program) 
PS Paurashava (municipal government) 
RCT Randomized Control Trial 
RDRS (Formerly Rangpur Dinajpur Relief Service) now the acronym is the name 
SDC Swiss Agency for Development and Cooperation 
SDLG Strengthening Democratic Local Governance (program) 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 2 
SOW Scope of Work 
TA Technical assistance 
TO Task Order 
TOR Terms of reference 
UGiiP Urban Governance Infrastructure Improvement Project 
UNDP United Nations Development Programme 
UP Union Parishad 
US Unnayan Shamannay 
USAID United States Agency for International Development 
UZP Upazila Parishad 
UzPAB Upazila Parishad Association of Bangladesh
V-AID Village Agricultural and Industrial Development Program 
VGD Vulnerable Group Development program 
VGF Vulnerable Group Feeding program 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 3 
EXECUTIVE SUMMARY 
This evaluation aims to provide USAID/Bangladesh with an informed assessment of the 
Mission’s Strengthening Democratic Local Governance (SDLG) project over its first 39 months 
from December 2010 through February 2014, at a cost of $11.6 million. SDLG covers 450 
township-size union parishads (UPs), 50 municipalities or paurashavas (PSs), and 100 county￾level upazila parishads (UZPs) for a total of 600 local government units (LGUs) Of the 600 
LGUs, 200 were to have partnered in previous USAID local government projects. 
SDLG has five program components. Component 1 and supported research, policy dialogue, 
and advocacy aiming to reform local government laws, along with introduction and 
demonstration of innovative practices in targeted LGUs. Component 2 aimed to improve the 
capacity of Bangladesh’s local government associations (LGAs), to improve their financial 
sustainability, their capacity to provide with training and technical assistance, and to advocate 
for policy reform. These two components consumed about 20 percent of SDLG’s budget. 
Component 3 intends to improve LGU transparency and effectiveness in service delivery in 
their ten compulsory functions, and in additional basic services. Component 4 adopts an 
existing model of “citizens’ forums” to promote participation in planning, budgeting and 
oversight. In SDLG these were termed Citizen in Government (CiG) groups, at the ward 
(neighborhood) level to set planning priorities, and channel citizen service needs to LGUs. 
Components 3 and 4 were implemented by nine partner non-governmental organizations 
(PNGOs). These two components absorbed 80 percent of SDLG’s budget. SDLG’s Component 
5 focused largely on including women and youth in local governance leadership roles. 
In implementing the different activities, SDLG was also tasked to maintain close contact with 
Government of Bangladesh (GOB) institutions, including the Local Government Division (LGD), 
“to ensure cooperation and buy-in from the GOB”. Reflecting the recommendations of 
evaluations of UAID’s earlier local government projects, SDLG was also to collaborate closely 
with other donor programs, particularly the WB’s LGSP. According to the Task Order, SDLG 
would provide demonstration sites and models: for this reason, it placed emphasis on “creation 
and dissemination of best practices”. Finally, it was also to work closely with USAID’s own 
programs in other sectors like health, environment, economic growth and food security. 
The objectives of the evaluation scope of work (SOW) are to assess SDLG’s actual results 
against targets; to gauge SDLG’s implementation tools and management in meeting project 
goals; and to make recommendations for future programs. There are six evaluation questions: 
 Effectiveness in achieving objectives; 
 Efficiency as the best approach, as opposed to alternative project approaches; 
 Management and administration in working with the PNGOs, other donors, LGUs, 
citizens, and the Government of Bangladesh (GOB);
 Sustainability after the end of the project (EOP); 
 Relevance to current development circumstances in Bangladesh; and 
 Cross-cutting impact in addressing gender and youth issues. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 4 
The evaluation team’s methodology comprised document review; key informant interviews in 
Dhaka; semi-structured interviews with individual key informants and with LGU and CiG 
groups during two field visits, and analysis of results of three surveys contracted by SDLG. 
Several LGU clusters were chosen for the team’s visits. Two constraints emerged: time did not 
permit visits to control LGU sites, and ongoing UZP elections meant excluding many areas from 
the sample LGUs. Also, elected officials in many of the LGUs visited had absented themselves 
to engage in electioneering, which limited the team’s findings concerning SDLG’s UZP activities.
FINDINGS FOR COMPONENTS 1 (Roles and Authorities of Local Governments 
AND 2 (Advocacy and Capacity Building of Local Government Associations) 
Effectiveness: SDLG focused its policy research effort mostly on Upazila issues, creating a 
new LGA (Upazila Parishad Association—UzPAB) and placing less emphasis on working with 
LGAs representing the UPs and paurashavas. Instead, SDLG concentrated on an advocacy 
endeavor to influence policymakers to implement the Upazila law. These efforts succeeded in 
getting pledges into the ruling party’s campaign manifesto, which have yet to be acted upon. The 
cutbacks in SDLG’s budget had a significant impact on the project’s activities in advocacy. 
Management and Administration: SDLG gave core support to the Municipal Association of 
Bangladesh (MAB) and the Bangladesh Union Parishad Forum (BUPF), while pressing them to 
increase their dues-collection and improve their internal organization. SDLG budget cutbacks 
foreshortened these activities, and caused conflicts with the LGAs. SDLG consulted with other 
donors, and with relevant GOB offices, but these contacts did not lead to involvement with 
either donor projects or the GOB. 
Sustainability of LGAs: SDLG’s emphasis on LGA membership dues bore some results, with
BUPF increasing its dues-paying membership from eight to 23 percent and UzPAB from nine to 
20 percent, as MAB experienced a slight growth from 41 to 43 percent. But while MAB will 
likely survive SDLG’s EOP and the termination of its support, BUPF and UzPAB (the Upazila 
Parishad Association of Bangladesh) expressed pessimism. 
Relevance: SDLG’s research priorities are relevant to UzPAB’s expressed needs, though less 
so to BUPF’s and MAB’s. There was no apparent link between the research and advocacy work 
in Components 1 and 2, and the field activities in Components 3 and 4. Again, progress in this 
direction was forestalled by project budget cuts, as SDLG preferred to maintain its field 
presence in Components 3 and 4, to the detriment of activities in support of policy research, 
advocacy, and collaboration with other donor agencies and with other USAID sector programs. 
CONCLUSIONS FOR COMPONENTS 1 AND 2 
Effectiveness: SDLG’s research and advocacy promotion engage with national policymakers, 
but this high-level lobbying does not appear to have thus far translated into policy change. The 
budget cutbacks had a negative impact on relations with LGAs and partner research 
organizations. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 5 
Management and Administration: SDLG managed well the relations with PNGOs and the 
fieldwork with partner LGUs. However, SDLG did not build partnerships with other donors in 
research or advocacy efforts, in particular with LGSP—which was a specific TO requirement. 
Nor did SDLG engage effectively with the GOB, in particular with the LGD senior bureaucracy. 
Sustainability: Owing to difficulties in reaching agreements with the LGAs, and then a major 
project budget cut, SDLG was unable to carry out the LGA capacity development activities, for 
example, to develop their capacity to provide training services for member local governments. 
And, while SDLG’s efforts to increase LGA dues-paying did produce more revenue, the 
improvement will probably not be enough to enable BUPF and UzPAB Secretariats to survive. 
Relevance: To attempt to promote policy reform in all three levels of local governance is 
relevant but overly ambitious. The research supported by SDLG was at best exploratory, 
summarizing and reiterating earlier research, without delving deep into LGU management 
issues. It failed to address effectively the TO requirements for documenting LGUs’ innovative 
service delivery solutions and disseminating them through studies and policy advocacy events. 
FINDINGS FOR COMPONENTS 3 Transparent and Effective Service Delivery by 
Local Governments), 4 (Citizen Participation in Local Decision Making) and 5
(Windows of Opportunity) 
Effectiveness: SDLG did well in meeting its operational targets for these two components, to 
judge by the project’s Performance Management Plan (PMP) reports. The survey results are less 
positive, indicating that in some cases (PS integrity in managing funds and citizen evaluation of 
public services in UPs and PSs) the control group LGUs showed more improvement than the 
treatment group. Most interviewees saw the training and technical assistance as useful. 
Efficiency (Alternatives): The main efficiency issue is the decision to work across three 
levels of local government and in such a large number of LGUs. SDLG was not able to address 
some of the Task Order requirements effectively by spreading budget resources so thinly. 
Management and Administration: The administration of the project field activities was 
adequate, and overall LGU officials were satisfied with the support. The PNGOs appear to 
have been effective in their role as training providers and facilitators. Yet, despite this good 
management of field activities, SDLG was unable to engage other donors or GOB agencies, to 
“upload” program innovations into other projects or into policy. 
Sustainability: Despite widespread enthusiasm for the program among the LGUs and CiG 
members, the team found little evidence that the CiG groups had built the capacity to continue 
practicing the skills they had acquired. This is not surprising, in the sense that SDLG’s LOP 
allowed only 12 to 18 months of full-fledged field implementation before it ended in December 
2013. Moreover, as observed above, the absence of linkages to other donor programs or GOB 
agencies reduced the potential dissemination and replication of its program activities. 
Relevance: SDLG’s interventions were relevant to the LGUs’ major “demand side” 
governance problems, related to weak accountability and responsiveness. However, SDLG’s 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 6 
capacity development, less than 10 days in duration, was unable to address complex service 
delivery management issues. 
Cross-Cutting Issues: In Component 5, SDLG ensured that all CiG groups included one￾third women member and sponsored several leadership training programs for women. As for 
youth, members under 25 years old constituted 12 percent of the overall CiG groups. 
CONCLUSIONS FOR COMPONENTS 3, 4 and 5 
SDLG’s Major Achievement: Combining Revenue Generation with Citizen 
Participation: Bringing together revenue generation and citizen participation was SDLG’s 
most noteworthy innovation. Activities to improve OSR were sufficiently attractive to lead 
LGUs to overcome their reticence in working with the CiG groups, and soon elected local 
officials came to see the CiGs as a “helping hand” for the local administration, by reaching down 
to the ward level. 
Low Sustainability of Project Interventions: SDLG’s startup left little time for field 
implementation– 18 months at most and generally less than that. This was simply not enough 
time to begin impacting local political culture, as the opinion surveys showed. Thus, even 
within the target LGUs, the evaluation team has doubts about sustainability of interventions. 
Project Design and Budgeting: Even without the cuts in the SDLG budget at the end of 
year 2, the Task Order was overly ambitious in requiring SDLG work in 600 LGUs. This meant 
that SDLG was able to address a narrow range of governance processes, and with a relatively 
low amount of training per LGU. In applying the budget cuts, SDLG opted to maintain this 
large set of LGUs, which kept the program from offering “demand driven” support to LGUs.
 
Policy Linkages: The fact that SDLG was ineffective in channeling its innovations into policy 
meant that the major project innovation in UPs and paurashavas (linking the CiGs to elected 
Councils to promote better governance) had no broader impact beyond the “demonstration”
LGUs. 
Add-on components: When an add-on feature like gender links directly to a mainstream 
USAID theme, it can be incorporated into new projects straightforwardly. Thus SDLG had no 
trouble including women’s participation in its programming for Components 3 and 4. 
RECOMMENDATIONS 
1. Concentrate Resources. The evaluation team recommends that USAID rethink the 
design of subsequent local government projects. USAID should prioritize, focusing on either 
rural LGUs (either UPs or Upazilas) or urban LGUs (paurashavas). Also, the project could 
use fewer treatment LGUs.
2. Ensure Continuity of Interventions. For any effort promoting change in political culture 
to build up sufficient momentum to sustain impact after EOP, the project must last longer 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 7 
than the 39 months of SDLG’s lifetime. A five- or six-year effort would be much better, and 
also ensure that successor local governance projects work in the same set of LGUs.
3. Build In a Service Sector Focus. The evaluation team recommends that the next 
project focus on decentralizing service delivery in one or more of these sectors. It should 
have sufficient resources for intensive and sustained capacity development for management 
improvements and support for citizen participation and accountability mechanisms to 
pressure LGUs to be more responsive. 
4. Incorporate Change Management Interventions. The focus of the local government 
project should be expanded to include piloting of “supply side” management solutions in 
specific service sectors. Work in Upazilas could incorporate change management and 
process reengineering approaches, working directly with the GOB line ministries.
5. Look at IT Solutions. In designing the next RFP, USAID should look at how UPs and 
Upazilas can improve how they manage between levels, for example through better 
collection and utilization of information. This could be coordinated with the GOB, perhaps 
through the PMO’s A21 or the Governance Innovation Unit. The project could look at 
using governance and management indicators to benchmark LGU performance.
6. Collaborate with GOB in Project Implementation. The discussion of “management 
solutions” in the previous recommendation suggests, too, that future local government 
projects need to dedicate more effort to engagement with the GOB, recognizing that 
government agencies have increasing fiscal, human and technical resources. The 
sustainability of project interventions will best served through direct involvement with the 
GOB, in both project design and implementation, to ensure buy-in from relevant ministries 
and agencies.
7. Engage/Coordinate/Collaborate More with Other Donors. The team suggests that 
USAID explore how to complement and extend the impacts of other donor programs in 
the field of local government. This may require much more preparation by USAID prior to 
issuance of the RFP, to better assess opportunities and reach agreements with potential 
partner programs on specific modalities and objectives of collaboration. 
8. Separate Contract Mechanisms for LGA Support. USAID should leave out LGA 
support/advocacy from a future project, and look instead for a separate contract mechanism 
under USAID Forward. These might include direct grants, conditional on pre-grant 
enquiries and appropriate capacity development prior to approval of funding. However, this 
is probably not the main priority, in the context of reduced budgets for DG programs. 
9. Continue to Use Rigorous Evaluation Approaches. The evaluation team 
recommends that USAID continue to use RCT approaches to evaluate the impacts of its 
interventions in strengthening local government. The team recommends that USAID 
support another application of SDLG’s survey of treatment and control LGUs, in 2015 or 
2016, to examine the sustainability and mutation of project interventions. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 8 
1. INTRODUCTION 
The project under review in this final performance evaluation, Strengthening Democratic Local 
Governance (SDLG), is the latest in a series of USG-sponsored decentralization initiatives 
stretching back to 1952 and the Village Agricultural and Industrial Development program (V￾AID) launched that year in what was then Pakistan. More recently there have been four local 
governance support efforts beginning with the Democracy Partnership (1997-2002), National 
Constituency for Strong Local Government project (NCSLG, 2001-2005), the Improving Local 
Level Governance Program (ILLG, 2002-2011), and the Democratic Local Governance Program 
(DLGP, 2005-2008). 
SDLG has been the largest in terms of total geographic coverage, providing assistance to 600 
local government units (LGUs). SDLG has also been the most complex of these local 
governance projects, including three levels of LGUs and a component focusing on nationwide 
local government associations (LGAs) and national policy reform. This report presents an 
evaluation of the 39 months from December 2010 to February 2014 of the SDLG program.1
SDLG’s goals and objectives can be summarized as a development hypothesis holding that 
strengthening the advocacy capability of LGAs at the national level can influence local 
government policy reform, which along with improving transparency and citizen participation at 
the local level will foster more responsive local governance and improved public service 
delivery. The project was implemented by Tetra Tech ARD of Burlington, Vermont, working 
with nine partner non-governmental organizations (PNGOs). 
2. THE DEVELOPMENT PROBLEM 
AND USAID’S RESPONSE
Throughout their history, local government in colonial Bengal/East Pakistan/Bangladesh have 
remained weak, underfunded, subject to bureaucratic control from above, unaccountable to 
their citizens, and manipulated as a patronage enterprise by the political party in power.2
Although the Constitution of Bangladesh calls for elected councils at all three tiers—union, 
upazila, and district—directly elected councils have been in place on a regular basis only at the 
union level since the country won independence in 1971. In the 1980s under the military 
government of General Ershad, upazila parishads (UZPs), or county-like units, which now 
number almost 500, began operating through directly-elected chairs (elections occurred twice), 
 
1
 In February 2014, SDLG was extended at a reduced level of funding until March 2015, but our SOW called for an 
evaluation of SDLG’s work in the first 39 months only, and accordingly, the evaluation team will not devote 
attention to the extension. 
2
 This interpretation has been abundantly documented over several decades, for example in Siddiqui (2000) and 
Khan (2009). 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 9 
but this system was abolished upon the return of democratic governance at the national level in 
1991 by a national government and parliament intent on maintaining central power. 
More recently, the Caretaker Government (CTG) of 2007-2008 promulgated an ordinance 
restoring Upazila Parishads (UZPs) with directly-elected chairs and, for the first time, two vice￾chairs, one of whom must be a woman. The Awami League (AL) that took power in 2009 
modified the ordinance so as to reduce the power of the UZPs, requiring them to take 
“guidance” from the Member of Parliament (MP). In addition, the government failed to establish 
UZP authority over the offices of the GOB agencies grouped together in the Upazila. Thus 
enfeebled, the UZPs have struggled to exercise any real power over Upazila management and 
service delivery. Elections for chair and vice-chair took place in 2009 accompanied by much 
enthusiasm, and at the time of this evaluation in 2014, new UZP elections were under way, in a 
context of intense inter-party competition and widespread violations of electoral procedures. 
The country’s 487 Upazilas are divided into 4,571 union parishads (according to the 2013 
Upazila Manual published by the Local Governance Department—LGD). Every union is 
governed by a UP, divided into nine wards, each electing a UP member; there also reserved 
seats for women UP members, representing three wards each, which brings the total number 
of councilors to twelve, plus one directly-elected chair. The present system has been in place 
since 1983 with minor changes in 2009. UPs have brought service delivery closer to the 
citizenry through UP “complexes” that house government workers in health, agricultural 
extension, and the like. But as in the UZP, these officials belong to their line ministries, and 
elected UPs have no authority over them. For their budgets, the UPs have been dependent on 
government grants, which increased somewhat over the last decade with the Local 
Government Support Program (LGSP), a World Bank initiative. UPs also generate OSR through 
taxes and fees, but these sources of revenue have been restricted in their scope. 
As in other countries, there are also urban bodies with their own jurisdictions and service 
responsibilities: paurashavas (PSs), of which there were 319 total at last count, as well as 11 
larger municipalities organized as city corporations. As befits urban areas, the paurashavas 
provide many more public services (e.g., water, sewage, roads, and streetlights) than the UPs, 
receive larger government grants, hire more municipal employees, and have more revenue￾generating power. But given the rapid urbanization now occurring in Bangladesh, the PSs are 
increasingly strapped for funds. They are governed by a directly-elected mayor and a council 
consisting of elected ward representatives.3
 
According to the SDLG Task Order, USAID designed the project to address the longstanding 
weaknesses of LGUs noted at the beginning of this section. SDLG was a relatively large project 
by USAID-DG standards: it worked in six of the country’s seven divisions; in 22 of its 64 
districts; 100 of its 480 Upazilas; 50 of its 309 municipalities; and 450 of its 4,500 unions. Of 
the 600 partner LGUs, 200 were to have participated in previous USAID local government 
 
3
 The number of wards varies according to a pourashava’s population, and thus, the number of reserved seats for 
women would also vary, though it equates to roughly one-third of the open seats. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 10
projects. At the same time, it provided support to three national LG associations (LGAs): the 
Bangladesh Union Parishad Forum (BUPF); the Municipal Association of Bangladesh (MAB); and 
the newly established Upazila Parishad Association of Bangladesh (UzPAB). 
SDLG’s budget was initially funded at $19.3 million over the life of the project (LOP). However,
the recent USG sequestration imposed a reduction of about 40 percent to $11.6 million at the 
start of Year 3 (January 2013).4
 This large budget supported four major program components, 
along with a fifth component to fund emerging “windows of opportunity.” The components fit 
together as shown in Figure 1 (see Annex V). A brief description of components 1 through 5 is 
provided below. A more in depth discussion of each component is included in the Statement of 
Work contained in Annex I. 
 Component 1 activities supported policy-relevant research, dissemination, and policy 
dialogue to increase LG roles and authorities. This included operational research on 
innovative practices in LGUs, especially in service delivery, and promotion of inter-LGU 
partnerships. SDLG emphasized research and advocacy to address gaps, ambiguities, and 
conflicts in the regulatory framework.5
 
 Component 2 activities develop the capacity of the three LGAs as membership bodies, 
as training providers, and in their policy advocacy. These activities were intended to link 
to SDLG’s Component 1 policy research and advocacy. Much of this work focuses on 
building LGA membership and dues collection and internal democracy through elections 
of the executive bodies. 
 Component 3 activities aimed to increase the transparency and effectiveness of LGUs 
through a combination of (1) training and TA for local officials and (2) the potential 
impact that the CiG groups created through Component 4 would have on elected 
councils and standing committees. 
 SDLG Component 4 activities supported increased citizen participation in LG decision￾making through the Citizens in Government (CiG) groups. In this way, there was a 
direct link with Component 3 activities. In practice, the CiG groups were the linchpin of 
Components 3 and 4. The CiG groups worked with SDLG to promote ward shavas to 
set planning priorities6
 and UP-wide open budget meetings, to support the LGUs’ work 
in tax assessment and collection, and to encourage better service delivery through 
direct participation in UP Standing Committees and direct contact with elected 
members.7
 
4
 Subsequently, $1.3 million was restored to SDLG’s budget in February 2014 for an extended phase to last until 
March 2015. The evaluation team did not include this extension in our evaluation, as per note 1 above.
5
 In particular, the Local Government Acts affecting UPs, PSs,and UZPs.(GOB 2009). For a summary of the acts, 
see Panday (2011b).. 
6
 The cover photograph for this evaluation report shows a ward shava meeting in process ion one of the SDLG 
partner Union Parishads 
7
 The CiG groups were formed initially in each ward of each UP and PS included in the SDLG project, when by 
consensus or a show of hands at a public meeting organized by SDLG’s implementing NGO 10 citizens were 
chosen. These sets of 10 people each then chose three among them to represent the ward at the LGU level. In 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 11
 The Component 5 activities sought to identify opportunities for innovations in 
transparent and participatory public administration in local government to support the 
other SDLG program components. This component focused largely on efforts to include 
women and youth in leadership roles in local governance. 
Components 3 and 4 cost far more to implement than 1 and 2. While Components 1 and 2 
together comprised 18 percent of total programming funds through December 2013, 
Components 3 and 4 consumed 82 percent. Funding for Component 5 was quite minor and 
was included in the other four Components. 
The Task Order aimed to contribute to the Mission’s Development Objective 1, “Citizen 
Confidence in Governance Institutions Increased”, its IR 1.4, “More responsive elected local 
government”, and Sub-IRs 1,4.1 and 1.4.2. By expanding the roles of local government, people 
will be able to locally determine their own priorities and by improving service delivery, they will 
also see that local governments are more effective—and in this way they will have greater 
confidence in their democratic institutions. 
3. EVALUATION PURPOSE & 
QUESTIONS 
3.1 EVALUATION PURPOSE 
The purpose of this evaluation is to provide USAID/Bangladesh with an informed assessment of 
the Mission’s SDLG project. USAID will use the evaluation to draw lessons for the selection, 
design, and implementation of future projects. The evaluation will help USAID to gain a better 
understanding of the sustainability of SDLG program outcomes and to plan for sustainable 
projects. 
3.2 EVALUATION QUESTIONS 
USAID identified three principal objectives for the evaluation, as outlined in the Scope of Work 
(SOW), which is provided in Annex I: These three objectives are: 
 Review the SDLG program’s overall performance by assessing actual results against 
targeted results; 
 Assess the efficacy of the SDLG implementation tools and management structure in 
meeting the objectives; 
 Make recommendations to USAID/Bangladesh concerning future programming in the 
local governance sector. 
 
this way a CiG central group of 27 (three including at least one woman from each of the nine wards) was selected 
for each UP. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 12
The SOW also sets six themes to evaluate through a series of questions. The areas of interest 
are: 
1. Effectiveness: To what extent has the SDLG program been successful in achieving its 
planned objectives? Were there any unexpected outcomes? What added value has USAID 
brought to the local governance sector in Bangladesh through SDLG? 
2. Efficiency: Is there evidence from the implementation of SDLG to suggest that alternative 
program approaches may have been more successful? 
3. Management and Administration: How effective and flexible has the SDLG 
management been in working with implementing partners, other donors (e.g. World Bank 
LGSP) and beneficiaries, such as local government units, locally elected officials, citizens, and 
GOB? 
4. Sustainability: How sustainable are SDLG program activities, and what measures could 
have been taken to enhance sustainability? 
5. Relevance: To what extent are the project’s objectives still relevant to the current 
development circumstances in Bangladesh, and will they provide sufficient guidance for 
appropriate programmatic and technical assistance decisions? 
6. Cross-cutting Issues: To what extent were gender and youth effectively addressed by 
SDLG’s interventions in the targeted areas?
4. EVALUATION METHODOLOGY 
AND LIMITATIONS 
The data gathering phase of this evaluation took place during February-March 2014. The team 
consisted of four members: Harry Blair, Visiting Fellow at the Political Science Department at 
Yale University and the Team Leader for this evaluation; William Cartier, Chief of Party for the 
BDGPE project and Evaluation Methodologist for the team; Akhter Hussain, Professor of Public 
Administration at Dhaka University and Senior Local Governance Specialist for the team; and 
Md. Naim Mostafa, Local Evaluation Specialist for BDGPE and the team. 
FIVE ELEMENTS OF DATA COLLECTION PLAN 
The team developed an Evaluation Design Matrix (presented as Annex VIII) around the six 
evaluation questions, listing the assumptions it was making about each question and the ways it 
would address each one in its field visits, Dhaka interviews and document review. The 
elements of the team’s data collection plan are presented below.
Document Review: The team reviewed SDLG’s project documents and reports to inform 
evaluation planning and background research. The project’s performance reports over the LOP 
were the most useful in this process, but the commissioned studies such as annual work plans, 
opinion surveys, Performance Management Plan (PMP) reports, and the SDLG task order were 
also very valuable to this component. The list of documents reviewed is provided in Annex II. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 13
Key Informant Interviews in Dhaka: In addition to relevant DG staff at the USAID Mission 
in Dhaka and staff at the SDLG office, the evaluation team interviewed GOB officials and staff of 
other prominent donors in the local governance field. The team was also able to meet several 
of the researchers involved with SDLG’s commissioned studies and interviewed a number of 
staff at the three LGA offices in Dhaka. A list of all key informants the team met with is 
available as Annex IV. 
Key Informant Interviews during Field Visits: The evaluation team took field trips to the 
Khulna and Rajshahi Divisions. On each trip the team interviewed elected officials, government 
servants, and CiG groups. A list of those seen is also part of Annex IV. 
Semi-Structured Interview Instruments: The evaluation team developed two open-ended 
questionnaires for its field visits and pilot-tested each during its first two days with field visits to
a UP, a PS, and a UZP. The protocols are provided in Annex III. 
 Individuals – LGU chairs for UPs, PSs, and UZPs; secretaries for UPs and PSs; and Upazila 
Nirbahi Officers (UNOs, the senior executive officer for an upazila). 
 Groups – elected council members, standing committees, Citizen-in-Government groups, 
women members of these bodies. 
Survey Data: The team also reviewed the three opinion surveys commissioned by SDLG, 
These three surveys, each employed a Randomized Control Trial (RCT) approach to include 
about 4,000 respondents in the treatment groups and almost 1,500 in the control groups: 
 A baseline study conducted in March-April 2011 before Components 3 and 4 began their 
actual field implementation; 
 A midline survey in December-January 2012-13; and 
 A final survey as work with the LGUs was coming to an end in December 2013. 
Following guidance from the Monitoring and Evaluation (M&E) office at the USAID Mission, the 
evaluation team only referenced the surveys to comment on the PMP measures employed by 
the SDLG project. 
SELECTION OF SAMPLE LGUS FOR FIELD VISITS 
The team organized the field work in geographical clusters of UPs, PSs, and UZPs, dividing the 
team in order to cover a larger number of partner LGUs. Because all four team members 
traveled for field visits, the team was able to split into two two-person sub-teams for data 
collection, except for the visit to the first UP/PS/UZP cluster where the team tested data 
collection tools and reinforced a systematic approach across sub-teams. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 14
In choosing the clusters of partner LGUs, the first step was to exclude clusters undergoing 
upazila elections on just prior to or during field work.8
 For the remaining clusters, the team 
chose at random using SDLG’s Local Government Capacity Index and the Score Cards 
(A/B/C).9
 The selected clusters included two Upazilas with higher-performing paurashavas and 
two Upazilas with lower-performing paurashavas. In each upazila, the team selected a union 
parishad (either a high-performing or low-performing UP according to the SDLG data) at 
random. Figure 2 (see Annex V) summarizes the final site selection. In one of the selected 
Upazilas (Sahjadpur in Sirajganj District), there was enough time to choose an additional UP.10
Figure 3 also located in Annex V is a map showing the field sites where the evaluation team 
gathered data. 
LIMITATIONS 
Small Sample Size: Due to time constraints, the team was able to visit only four SDLG 
Upazilas, and these four Upazilas may not have been representative of all SDLG’s sites. To 
obtain a truly valid sample, the team would have had to include all three levels of LGUs and 
areas supported by all nine project implementing NGOs. This implies a 27-cell matrix (3 x 9 = 
27) to just have a single entry in each cell – an effort far beyond the level of effort (LOE) 
available. So, based on these limitations, the team designed a sampling scheme that it believed
provided a reasonably complete picture of SDLG. In other words, while the evaluation team 
cannot say that the sample size was statistically representative of the universe of SDLG LGUs, 
they believe that the sampling design found enough variation to be illustrative of that universe. 
Selection Bias: As noted in Figure 3 (see Annex V), the team’s field visits were clustered in 
Rajshahi and Khulna divisions, suggesting some selection bias, given the lack of geographic 
diversity and the fact that the sites were chosen based on geography and whether or not they 
had an ongoing election (rather than randomly). This should not impart any serious bias since 
these two divisions included more than half of the total SDLG-supported district sites (12 of 22 
sites), upazila sites (56 of 100 Upazilas), union sites (253 of 450 unions), and municipalities (30 
of 50 municipalities). 11
 
Upazila Parishad Election Campaign Interference: As previously mentioned, the team’s 
first filter for choosing field visit sites was to eliminate all project Upazilas where campaigns 
 
8
 The electoral turnover rate among Bangladeshi LGU chairs and mayors has been quite high in the past. Many 
chose not to contest, and large numbers were defeated at the polls. It thus seemed reasonable to assume that the 
same could likely be true for UZP chairs and vice-chairs. 
9
 SDLG developed an elaborate index to gauge LGU capacity but then when project funds shrank found it 
necessary to switch to a simpler score card system. 
10
 It should also be noted that one of our selected pourashavas (Jibannagar) is crossed out and another 
(Chuadanga) is substituted. This was because we found that the mayor, a Jama’at member, was in hiding at the time 
of our visit, so we selected another PS in the same district. 
11
 It is virtually impossible to completely eliminate selection bias in any kind of sample, but the team believes it 
has done its best to do so in this evaluation, given the context of cost and time constraints. For a couple of good 
surveys of the problems involved, see Vella (1998), also Collier and Mahoney (1996). 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 15
would be in progress for the five-phase elections taking place. As it turned out, this “filter” did 
not fully filter election interference, as elected officials in non-campaigning areas were in many 
cases deeply involved in nominating candidates, campaigning for (or against) candidates in other 
constituencies, and lobbying with party officials visiting from Dhaka. Thus, the team was unable 
to meet with several UZP chairs, PS mayors, UP chairs, LGU council members, and even 
UNOs. In some cases it was possible to meet with vice-chairs or acting mayors, which was 
helpful, but not nearly as useful as meeting with the actual officials in charge of the LGUs. 
 
5. FINDINGS AND CONCLUSIONS 
The findings and conclusions are presented in two sections: one for the national-level 
Components 1 and 2 and the other for its local-level Components 3 and 4. Each section follows 
the SOW’s evaluation questions, presenting the findings in response to all the questions and 
then the conclusions stemming from those findings. Recommendations drawn from findings and 
conclusions are presented separately at the end of the report. 
To help manage the array of questions and answers, Figure 4 in Annex V clearly states the 
essence of each question and the team’s answers for Components 1 and 2 and Components 3 
and 4. In perusing this section of the report, reviewing the PMP indicators located in Annex VI
will provide useful context for the reader. At the end of the section, the report addresses
Component 5, which focuses on gender and youth inclusion in the program. 
5.1 FINDINGS FOR COMPONENTS 1 & 2 
SDLG’s Component 1 objective is to support advocacy initiatives for reforms to expand roles, 
authorities, and resources for local governments. The objective of Component 2 is to increase 
the capacity of local government associations to advocate and lobby on behalf of their member 
LGs for policy changes to empower local governments and advance democratic 
decentralization. Thus, SDLG provided support to the two existing LGAs (BUPF for UPs and 
MAB for paurashavas) and assisted in the creation of UzPAB representing UZPs. 
5.1. Effectiveness: Achieving Results 
Effectiveness in Integrating Components 1 and 2: The team’s evaluation of SDLG 
effectiveness in implementing Components 1 and 2 begins with the SDLG Task Order 
requirements, which state, “BUPF and MAB should take the lead in advocating for reform,” and 
“new opportunities should also be explored to foster legal and policy reforms separate from 
the MAB and BUPF activities.” The first finding is that SDLG was not effective in integrating 
these components so that MAB and BUPF were engaged with the policy research agenda and 
able to independently take up the findings and use them for their own advocacy activities. 
Rather, both SDLG staff and MAB and BUPF officials told the Team that the project mainly 
focused on the “new opportunities” and gave priority to upazila issues. SDLG explained that 
this happened in part because of budget cuts, which required them to prioritize actions. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 16
SDLG partnered with two research organizations: Unnayan Shamannay and the Center for 
Urban Studies (CUS). Researchers interviewed from these organizations stated that the LGAs 
did not have a role in defining the research agenda. They said that both the research products 
and policy dialogue activities were presented as an initiative of SDLG’s/USAID’s. This was 
confirmed by officials from all three LGAs, who explained that they did not have input into the 
research agenda or the TORs. They were invited to the policy activities but did not have an 
important role in the events.12
BUPF has an eight-point agenda which highlights the tendency of the GOB to curtail UPs’ 
existing authorities through “unlawful circulars.” BUPF officials interviewed said that most of 
their issues were not incorporated into the SDLG policy advocacy activity. However, SDLG 
notes that three of the BUPF demands were incorporated into the project’s advocacy work. 
The management of SDLG also recognized that cuts to the project budget in 2013 created 
conflicts with the partner LGAs, BUPF in particular, which undoubtedly contributed to BUPF’s 
negative perceptions. 
MAB, for its part, has a 16-point reform agenda. A MAB officer stated that SDLG support did 
not advance this policy agenda further. The evaluation team’s review of the program research 
and activity reports confirms that SDLG did not take up these policy agendas. The SDLG 
management said that the difficult relationship with MAB caused difficulties in agreeing on a 
common policy advocacy strategy. Another key informant from one of SDLG’s partner 
research organizations was critical of the SDLG research effort, saying, “SDLG ignored the 
work already done by others,” in this case referring to MAB’s own efforts with the Local 
Government Commission under the CTG, and subsequent advocacy initiatives under the AL 
government post-2008 in the framework of the Bangladesh Urban Forum. 
Effectiveness of Linking Research to Policy: Under the rubric of research, the SDLG 
Task Order required a range of activities including research on LG reform; investigation into 
decentralized service delivery and in particular, health and education; support for innovative 
practices; and partnership building among targeted LGs. A review of the research products 
shows that SDLG’s research and advocacy activities were focused on clarifying “gaps,” conflicts, 
or obstacles to full implementation of the three LG Acts passed by the CTG and later amended 
and ratified by the AL government in 2009-2010.13 This focus on implementation of legislation 
was confirmed in the evaluation team’s interviews with SDLG staff.
 
12 A review of the various SDLG outputs (policy research reports, advocacy materials and policy dialogue report) 
supports the views expressed by the researchers and LGA representatives. In the national policy events 
themselves, the proceedings were introduced by SDLG; the moderators were from the research organizations, 
and the research findings were presented by the researchers. The LGAs had the role of commentators or 
panelists. The regional policy events, however, did give a higher profile to LG representatives, who were given 
opportunities to make presentations on LG issues, but LGA representatives were not included at that level either. 
13
 The three pieces of legislation are: the Local Government (Union Parishad) Act 2009; Upazila Parishad Act 2009; 
and the Pourashava Act 2009. See GOB (2009) and Panday (2011b). 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 17
Despite the TO requirement that SDLG document LGU practices “that will demonstrate their 
improved ability to provide services to their constituents”, the research products showed no 
in-depth research into service delivery. In SDLG reports and interviews with SDLG staff, 
Unnayan Shamannay, CUS, MAB, BUPF, and UzPAB, the evaluation team found no operational 
research to demonstrate or document innovation in service delivery and to take the findings 
into policy.14 One MAB official said, “CUS did some initial fact-finding in paurashavas, [but] after 
a long time passed there were [no] results, summaries, or products that would be useful for 
us.” According to BPF and MAB, the research done by SDLG and Unnayan Shamannay on UPs 
also reiterated the findings of other research, with the important exception of an audit and 
accountability study.15 SDLG notes that the budget cuts in 2013 limited its progress in this area. 
MAB representatives stated that SDLG was “too focused” on the Upazila issues and “didn’t give 
time to the municipal policies” or to support initiatives directly with the LGD on urban issues. 
One partner research organization observed that SDLG was not integrated into ongoing policy 
initiatives on urban governance such as the Bangladesh Urban Forum or the ABD funded UGiiP 
project with paurashavas. The same organization also noted that municipal governance and 
management is a technical policy field and requires in-depth, focused research and engagement 
with the central government. This organization further observed that SDLG in spreading its 
resources across three levels of LG, “could only scratch at the surface” of the issues.
UzPAB representatives consider that SDLG and Unnayan Shamannay were effective in getting 
their priority issues to decision-makers, using both public forums and closed door meetings 
with ministers and MPs. Reviewing the SDLG reports, the evaluation team concurs with this 
view. The founders of this consulting group are senior opinion-makers and gave SDLG access 
to policymakers at the national level. The Unnayan Shamannay researchers claim that SDLG 
was able to have the Awami League party incorporate elements of one of the SDLG policy 
briefs on decentralization into its January 2014 parliamentary election manifesto, which states, 
“More authority and responsibility will be delegated to the Zila Parishad, upazila parishad, and 
union parishad through democratic reorganization of the present centralized administrative 
structure.”16 However, at time of writing, according to key informants interviewed, this 
statement had not yet translated into new policy reforms on local government. 
GOB representatives and donor agencies questioned the SDLG approach to advocacy on 
upazila issues. Two senior local government officers interviewed stated that the upazila 
legislation had created “conflict in the bureaucracy” that would take decades to work out with 
each Ministry having to rethink its processes. Further, one officer stated, “Trying to force it 
through by transferring the revenue fund won’t help matters.” Representatives of two donor 
 
14 Components 3 and 4 did develop a highly innovative approach to improving service delivery with its combination 
of revenue generation and citizen participation, as noted later in this report. 
15 Sobhan, et al. (2013).
16 It was not possible for the evaluation team to establish attribution for the AL decision on the election manifesto. 
While it is true that SDLG did engage with senior GOB and AL officials in 2012 and 2013, so did many other 
donor programs. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 18
agencies echoed this view, noting that fully implementing the existing upazila legislation would 
need “close engagement” with LGD and the line ministries. 
Unexpected Outcomes: The most salient unexpected outcome of the research and 
advocacy was the conflictive relationship between SDLG and the two existing LGAs during the 
first and second years of the project. This conflict made it difficult for SDLG to reach an 
agreement with BUPF and MAB on the grants, which also impeded effective integration of the 
LGAs into SDLG’s research and advocacy activities. The situation as aggravated by the SDG 
budget cuts, which were felt more strongly in Components 1 and 2 activities. 
Value Added: In Components 1 and 2, the UzPAB executive leader and the Unnayan 
Shamannay researchers told the evaluation team that SDLG had brought Upazila issues to the 
fore at a time when the GOB was not interested in public dialogue on how to advance the 
decentralization agenda. The research products and advocacy outputs were unique and 
timely—only one other donor organization, UNDP, was conducting research on questions 
related to Upazilas. And, only SDLG was carrying out advocacy on these issues (UNDP worked 
directly with LGD). While these efforts have not yet resulted in policy reform, they do appear 
to have provided some value that could later result in policy reform.
With respect to LGA strengthening, despite the difficulties with LGA leadership, SDLG did add 
considerable value, in a variety of ways. It pressed upon the association leadership the need for 
sustainability through better communications and outreach to members, and through internal 
democracy. It also produced some innovative concepts for association strengthening, for 
example, the proposal for a communications strategy via cel phones (as all LGU leaders use cel 
phones intensively).17
 However, owing to budget cutbacks, in 2013 SDLG was not able to 
advance further in this direction, 
5.1.2. Efficiency: Alternative Approaches 
LGAs as Trainers: BUPF officials were critical of SDLG management, saying that they did not 
understand why SDLG “continued support for NGOs” but cut off support to the LGAs. The 
officials pointed out that early in the SDLG program they asked to be considered as a service 
provider for the training to UPs under Component 3 of the program. BUPF had recently 
received funding from the USAID-funded PROGATI program to train 25 “master trainers,” and 
they proposed to use these same trainers in SDLG. BUPF officials said that SDLG declined, 
arguing that BUPF did not have the necessary technical and management capacity. SDLG 
managers reiterated this view about BUPF in the interviews with the evaluation team. 
Coalitional Strategies: Several of the ongoing policy initiatives on decentralization in 
Bangladesh involve coalitions. For example, the Governance Advocacy Forum (GAF) led by the 
Wave Foundation is a coalition of 30 NGOs which addresses decentralization issues and which 
 
17 See SDLG. 2013. “Digital & Email Strategy for LGA Promotion. Strengthening Democratic Local Governance 
(SDLG) Project”. Dhaka: Unitrend Ltd. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 19
existed at the time SDLG was starting program activities. In the urban sphere, another policy 
initiative is the Bangladesh Urban Forum (BUF), initiated in 2011, with GOB ministries and 
departments, donor agencies, national and international NGOs, universities, and MAB. In 
reviewing SDLG advocacy activities, there is no evidence of involvement with these initiatives. 
5.1.3. Management and Administration 
SDLG Partners: Early on in the SDLG program, SDLG management and USAID made it 
known to the LGAs that continued support would require LGA leadership’s commitment to 
internal democracy through elections; increased membership; and increased revenues through 
membership fees. The MAB and BUPF executive board members describe this policy decision 
as “abrupt,” but they said that they agreed to the conditions and began to take the measures 
required by SDLG. For their part, SDLG management said that the two partner LGAs were 
slow to respond to the funding conditions, in particular, MAB, which took more than a year to 
organize its internal elections. SDLG management also emphasized that “this strategic emphasis 
[on internal democracy and sustainability] ran counter to the centralized nature of the existing 
associations and…raised the prospect of an end to their heavy reliance on donor funds”.18 
SDLG management notes that despite the delays, SDLG continued to provide core support to 
the LGAs to cover their secretariat costs and also paid for internal elections, regional and 
national conferences, and advocacy events involving the LGAs. SDLG also provided training to 
BUPF and MAB on board leadership, membership building, and dues collection. Some of the 
training with MAB was suspended because of lack of commitment. SDLG management also said 
that the delays in meeting the requirements and the difficult relations with BUPF and MAB 
leadership delayed considerably the negotiation and approval of grants to these LGAs. 
For their part, MAB and BUPF officials are critical of both SDLG and USAID handling of the 
relations with the LGAs. They appreciated the core support provided by SDLG for the 
operation of the secretariats. The problem was the support for program activities, especially 
advocacy and the establishment of training programs. One official stated, “We understood the 
new conditions, but we didn’t understand why the grant negotiations were delayed so long—it 
is as if they didn’t trust us.” He pointed to the fact that after almost two years of constant 
meetings on the terms of grant support, SDLG announced to them that it was removing the 
program activities from the proposed grants. SDLG project staff explained that the decision was 
made in response to the funding restrictions caused by the broader USG sequestration issue. 
Other donors: The SOW for this evaluation asks with respect to project management, “How 
effective and flexible has SDLG management been in working with implementing partners, other 
donors and beneficiaries…and GOB?” The TO stated, “Donor coordination is especially 
important as USAID will through this SDLG project work with a relatively small subset of local 
governments,” (page 7). It also said that collaboration with the major WB funded LGSP project 
 
18 This clarification comes from SDLG. 2014. “Comments of the Performance Evaluation of the SDLG Project from 
the SDLG Team”. Dhaka: SDLG. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 20
“is essential.” A Secretary from the LDRG&C Ministry and the management team of LGSP 
reported that they had not met with SDLG to coordinate activities; rather, according to the 
Secretary, SDLG limited its interaction to just “sharing” its own plans for work with the UPs
with LGD, without proposing any joint activities. SDLG management told the evaluation team 
that they had not attempted to coordinate their activities with LGSP; they portrayed the World 
Bank project in negative terms. 
With respect to other donors, SDLG managers pointed to the early initiative to create a 
“friends of SDLG” group, but this initiative did not bear fruit. Subsequently, SDLG participated 
in the local consultative group on LG, in which it reported on its activities. The bilateral 
agencies interviewed said that there were no concrete proposals from SDLG for collaboration. 
An ADB official said they were open to cooperation with smaller donor programs, “But this 
would take long-term planning and coordination—it couldn’t be done at the spur of the 
moment.” He said “It would need to be agreed with the project management teams,” referring 
to GiZ’s work on governance practices within the larger ADB UGIIP project and UNDP’s 
applied research work within the framework of the LGSP as examples of effective cooperation. 
The sole example of successful SDLG collaboration with other donor agencies was the efforts 
to link the SDLG partner LGUs to the World Bank-funded Horizontal Learning project. The HL 
project accepted five partner Upazilas and their respective UPs into the program, with purpose 
of disseminating their innovations with CiGs. This happened almost three years into the 
project, however, and will require several more months before it comes to fruition. 
Government of Bangladesh: Policy research by Dr. P. K. Panday contracted by SDLG in 
2011 argued for engagement with the GOB to resolve issues in implementation of the Upazila 
Law, recommending that “Advocacy is necessary with the LGD for taking an initiative to sit 
with the line ministries for issuing a circular directing their field level officials to send all files to 
the UZC for approval .”19 This conclusion of course repeats the injunction in the Task Order 
for SDLG to engage effectively with national policy-makers. However, SDLG’s advocacy work 
was more focused on legislators and party leaders than the bureaucracy. The evaluation team 
asked the LGD secretaries whether SDLG had provided opportunities for working directly with 
the LGD or other departments —both replied they had not interacted with SDLG. When 
asked whether there had been follow up meetings with SDLG one Secretary replied, “Not to 
my knowledge. We don’t know what the USAID project is doing with the Upazilas.”
5.1.4. Sustainability of LGAs 
Project Activities: With respect to Component 2, as Table 1 (see Annex V) shows, SDLG 
made significant contributions to the three LGAs, covering a large proportion of their core 
costs (secretariat) and funding project costs with the LGAs (internal elections, policy events, 
meetings, etc.), as well providing small grants for LGA activities. In addition, SDLG spent an 
estimated $199,567 for staff working with the LGAs. 
 
19
 See Panday (2011a: 90). 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 21
As a result, and despite the difficult relations between SDLG and the LGAs and the complaints 
from the LGAs about the change in the rules of the game, the team found that SDLG did 
contribute to LGA internal democracy; the higher rate of LGU membership affiliation and 
increased financial sustainability. Over the last year of the SDLG project, the rates of LGU 
dues-paying membership in the three LGAs increased from 41 percent to 42.9 percent in MAB; 
from eight percent to 22.7 percent in the case of BUPF; and from nine percent to 20.3 percent 
in UzPAB, according to data assembled by SDLG. The UzPAB and BUPF presidents recognized 
that the improvements in membership affiliation and dues collection were due to SDLG 
assistance through training for association capacity development. 
From interviews with the three LGAs, the evaluation team heard that MAB had taken actions 
to ensure its continued survival by obtaining free office space in an LGED building and raising its 
membership dues. MAB was also finishing implementation of a grant from DANIDA and was 
looking for additional opportunities for small grants. MAB officials expressed confidence that it 
would be able to make the transition from USAID core support, albeit with “some difficulties.”
In contrast, both BUPF and UzPAB state that they anticipate difficulties in maintaining their 
secretariat functions, without which it will be impossible to retain their membership. The 
UzPAB president, too, was pessimistic about the future of the association, saying, “SDLG made 
some progress in identifying the main problems, but the support should continue for another 
four or five years. We will try with our own resources but it will be very difficult.”
The evaluation team also asked SDLG about sustainability with respect to training provision by 
the LGAs and the GOB. The SDLG management team said that neither the LGAs, nor the 
GOB had the capacity to provide quality training for a large number of local governments in the 
relatively short time period of the project. Thus, SDLG’s decision was to design its own training 
program and to train partner NGOs to implement it. 
5.1.5. Relevance
SDLG was effective with UzPAB in linking research and advocacy to this LGA’s express 
priorities. US’s research focused on the implementation of two important provisions of the 
Upazila Act: the effective transfer to Upazila Council control over 17 GOB departments, 
including the spending of development and revenues funds; and the collection of land taxes 
directly by the upazila. The choice of these issues was supported by the UzPAB executive body. 
The other LGAs did not question the relevance of SDLG’s policy advocacy, but instead they 
wanted a closer fit with their own policy agenda and support for their own advocacy activities. 
The evaluation team’s review of the SDLG research products compared to other research on 
LG issues shows that in union parishad and municipal governance, the research and advocacy 
activities tended to focus on a set of general issues that have been amply discussed in past 
years. The SDLG project managers and the Unnayan Shamannay and CUS researchers all 
consider these issues still relevant. The donor agencies agree that decentralization and local 
government capacity development continue to be critical priorities. However, several donor 
officials suggested that working on upazila issues would require a sectoral approach. For their 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 22
part, LGD officials suggested that to be more relevant (and effective), USAID’s capacity 
development programs have to engage more in the technical, financial, and administrative details 
of a broader range of LG governance and management topics — the “nitty gritty,” as one LGD 
Secretary termed it. From the perspective of LGD, the SDLG’s thematic focus on revenue 
enhancement and participatory planning was too narrow, especially in paurashavas and Upazilas, 
which compared to UPs face a wider range of challenges in governance and management. 
5.2 CONCLUSIONS FOR COMPONENTS 1 & 2 
5.2.1 Effectiveness of Integration of Components 1 and 2 
In the case of BUPF and MAB, SDLG did not involve the associations sufficiently in the 
definition of research priorities or in the use of the research in function of LGAs’ advocacy 
activities. In the case of the Upazilas, SDLG and Unnayan Shamannay did incorporate some of 
UzPAB’s priorities, especially the need to eliminate MP interference and to give Upazilas 
management control over their funds. 
1. Effectiveness of Linking Research to Policy 
SDLG mainly focused on implementation of the three LG laws. The interviews with the LGAs 
and the review of SDLG documents suggest that SDLG gave priority to Upazila issues in its 
research agenda and advocacy. While senior GOB bureaucrats, donor agencies, and even 
SDLG's own researchers say that it would be productive to engage directly with LGD and 
other GOB ministries, SDLG did not take this route, involving LGD only tangentially at best. 
While effective decentralization requires clarity on “roles and responsibility,” it also depends on 
finding a way to manage between levels of government. For this reason, other donors are 
working directly with LGD on the implementation of the UP, paurashava and Upazila laws. 
SDLG's research and advocacy ignored this aspect, including the possibilities for working with 
the GOB within ongoing, multi-donor initiatives, which would have provided ample room to 
focus on improving specific aspects of LGU management, including service delivery innovations. 
2. Management of Research and Advocacy 
SDLG did not carry out effective research and advocacy activities in two of the tasks required 
by the TO under Component 1 Task C – Innovative Practices. SDLG failed to document and 
disseminate innovations for policy advocacy and reform; and Task D (Partnership Building), 
where the activities supported by SDLG were at time of writing still incipient. While SDLG 
argues that this is because of budget cuts in 2013, the evaluation team saw no progress towards 
these two TO requirements prior to the budget cuts coming into force. 
By opting for a high-level lobbying strategy with Unnayan Shamannay, SDLG was able to get its 
foot in the door with senior officials. In the absence of an immediate decision by government, 
and also now absent continued funding for Unnayan Shamannay, however, the momentum 
appears to have been lost. Nor did SDLG build partnerships with donor organizations in its 
policy research and advocacy activities, which might have carried forward the SDLG initiatives. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 23
Neither SDLG nor Unnayan Shamannay engaged other donors in the design of any of the policy 
research and advocacy strategies, nor did they coordinate with other, large-scale projects 
funded by multi-lateral institutions, in particular LGSP, as required by the TO. 
Referring to union parishads, SDLG's interventions ignored the challenge and opportunity 
posed by the impending conclusion of the WB’s LGSPII. The evaluation team considers that this 
was a critically important, missed opportunity for SDLG to engage with LGD, the LGSPII 
management team, other donors, and BUPF on making block transfers to UPs a permanent 
policy of the GOB. 
4. Sustainability 
SDLG's approach to training, and its reluctance to use the BUPF or other LGAs for the delivery 
of training, was based on effectiveness, not sustainability considerations. The trade-off was 
evident—SDLG was effective in using the partner NGOs to deliver training, but this training 
provision is not sustainable. These NGOs may well return to LG issues but only if another 
project provides the necessary funds. At a time when several other important donors in 
Bangladesh are withdrawing from local governance programs, this may not happen soon. 
While MAB will continue to survive, BUPF and UzPAB face an uncertain future; BUPF did not 
prepare for the day when USAID support would end, and UzPAB is still an incipient 
organization. Again, all of the LGAs expressed surprise and dismay that USAID would not want 
to ensure the sustainability of its investments in developing their organizational capacities. 
On the question of the sustainability of advocacy, SDLG’s main partner, Unnayan Shamannay, 
conducts research and advocacy according to the client’s wishes. It has no mission or mandate 
to continue efforts in this direction. CUS will continue to work in the area of urban governance 
and management. The SDLG initiatives did not connect with any other advocacy initiatives led 
by Bangladeshi organizations, such as the GAF or the BUF. Progress in this sense was cut short 
by the SDLG budget reduction, which gave preference to Components 3 and 4 activities. 
5. Relevance 
After reviewing the SDLG project documents, research products, and advocacy activity reports, 
the evaluation team concludes that the SDLG TO design was overly ambitious in attempting to 
support high quality, operationally relevant, and durable policy change interventions across all 
three levels of local government. From the interviews with LGAs, LGD officials and donor 
representatives, the team concludes that while the issues addressed by SDLG are relevant in 
principle, in remaining at a general level, without engagement with the partner LGAs and GOB 
agencies, they do not generate specific proposals that could be taken up by policy actors. 
The work by SDLG and its partner Unnayan Shamannay on Upazila issues was the most 
relevant of the advocacy activities and continues to be so. Most of the key informant interviews 
with donors and government officials highlighted the question of how to decentralize service 
delivery at the Upazila level as one of the most critical aspects of government effectiveness in 
Bangladesh today. However, from the interviews with GOB and donor agencies, the evaluation 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 24
team concludes once again that SDLG lacked engagement with government agencies (LGD and 
other ministries) on this topic. By supporting the immediate transfer of funds to upazila control, 
without working through the details of service delivery in each sector, LGD’s senior managers
saw SDLG’s activities as reflecting UzPAB’ s interests but not necessarily favoring a rational, 
planned process of policy change. For the senior bureaucrats in the Bangladesh civil service, 
such a wholesale transfer of power – in their words a “big bang” approach – is anathema. 
FINDINGS FOR COMPONENTS 3 & 4 
1. Effectiveness
1a. Achieving Objectives: The objective of Component 4 (“citizen participation in local 
decision making”) was “to ensure transparency and accountability in public decision making 
[that would] lead to improved service delivery,” while that of Component 3 (“transparent and 
effective service delivery by local governments”) was “to enable [local governments] to be 
effective and transparent service providers at local level.” Accordingly, the team interprets 
Component 4 to be aimed at producing the mechanisms that will (with additional support from 
SDLG) help bring about improvement in Component 3. More participatory government will 
support better service delivery.
For Component 4, the PMP data in Annex VI show SDLG establishing CiG groups in all 500 
LGUs (UPs + PSs) and participatory planning/budgeting implemented in 490 of them (PMP 
4.1).20 Moreover, SDLG met the LOP target of supporting five local mechanisms supported for 
citizens to engage local governments (PMP 4.4). For the two opinion questions (PMP 4.2), 
findings are somewhat mixed, though both show gains. Tables 2 and 3 (Annex V) show modest 
increases in citizen awareness of LGU revenue raising efforts for UPs and PSs respectively, 
while Tables 4 and 5 show a larger growth in citizen trust in LGU integrity. In the latter PS case, 
however, the “yes” answers were also improving, even more in the control municipalities (from 
16 percent to 58 percent answering “yes”) than in the treatment towns (17 percent to 57 
percent), implying that the “treatment” was irrelevant, or at least that other factors operating 
in both groups overshadowed the SDLG interventions in the treatment group.21
 
For Component 3, SDLG or exceeded all three PMP coverage targets (numbers of LGUs 
receiving USAID assistance in increasing capacity, number receiving assistance with revenue 
generation, and numbers of individuals receiving USAID training – PMPs 3.1, 3.2, 3.3). Revenue 
generation did go up, by 217 percent in the treatment unions as against 122 percent in the 
control units, while the treatment municipalities registered a 117 percent increase vs. a decline 
of 19 percent in the control paurashavas. 
 
20 The LOP target for participatory planning/budgeting (PMP 4.3) is shown as 600, which would include all 100 
UZPs, but SDLG wasn’t working along these lines in the UZPs. Therefore, the team assumes there must have been 
a typographical error here. 
21 Interestingly, the endline survey showed in a logistical regression that whether or not a respondent was in the 
treatment or control group had virtually no effect in predicting whether he/she had confidence in local elected 
officials (Sabet and Haque, 2014: 51). 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 25
The increased OSR generated by SDLG provided considerable discretionary income to be 
allocated by the LGUs. Data on how that money was spent in SDLG’s PSs and UPs is presented 
in Table 6. The largest single item is “Development Projects,” which represent service delivery, 
and likely some of the “Office, et al.” category would include service delivery as well. But the 
“Salaries” line indicates that a good portion of the additional resources went to increasing the 
pay of locally elected officials, particularly in the UPs, where aside from the secretary they are 
the only officials on the payroll.22
 
Satisfaction with public services proved a somewhat different story, as becomes clear in PMP 
indicators 3.5a and 3.5b (citizen satisfaction with service delivery in UPs and PSs respectively), 
shown in Tables 7 and 8 (Annex V). UP respondents, asked about quality of roads, answered 
“good” or “excellent” 53 percent of the time in the baseline survey, a number that climbed up 
to only 57 percent in the final survey, far below the targeted 30 percent over baseline (the 
control group registered a slightly larger increase, from 48 percent to 53 percent). In the 
paurashavas, citizens were asked about marketplace quality. The baseline-to-final “good” or 
“excellent” answers improved from 49 percent to 55 percent, this time well below the increase 
among control group respondents, which went from 58 percent to 80 percent. The greater 
increase within the control LGUs implies that public services improved more outside of the 
SDLG project area than within it.23
 
In its field visits, the team found that CiG groups were formed in every UP and PS (PMP 4.1). 
Respondents from LGU councils and CiG groups as well as chairs/mayors and LGU secretaries 
reported uniformly that the CiG groups were instrumental in mobilizing attendance at the ward 
shavas and open budget meetings (PMP 4.3), and further that the CiG members themselves 
played an active role in the discussion at those assemblies. Finally, all respondents said that the 
PNGOs had provided training and TA relating to the ward shavas, open budget meetings, 
standing committees, and LGUs themselves (PMP 4.4). 
All the LGUs visited reported receiving SDLG assistance in capacity building (PMP 3.1) and in 
particular, receiving assistance in increasing their own source revenue (PMP 3.2) – indeed, this 
was the most important aspect of the whole project in the eyes of the vast majority of 
respondents, as will be discussed in more detail below. And, all those the team talked with 
received at least some training from the PNGO project officers (PMP 3.3). 
Looking at SDLG’s work with the Upazila level, upon reviewing the program documents, 
reports and PMP, and in the interviews with SDLG management and PNGOs, it was found that 
SDLG was unable to roll-out its training for the targeted Upazilas until Year 3; indeed, the 
 
22 In addition to paying elected officials, the PSs employ a number of workers to provide services, so the “Salaries” 
line may well have included them also.
23 The small size of the control group in the pourashavas (n=44 in Table 8) may well be a factor here. It should be 
noted that in the baseline survey, 70 percent of respondents answered the question in both treatment and control 
groups, but in the endline survey, it was only 28 percent and 37 percent, respectively. Other factors also may have 
been at work here. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 26
design and piloting of the training materials for what SDLG termed “the most contentious and 
complex” of the three levels of local government was not completed until the end of Year 2. 
SDLG was not able to look at the impact of its work on citizen satisfaction with Upazila 
services at all—which is unfortunate, as the Upazila is the most important service delivery 
organization in the GOB. The SDLG Upazila surveys were limited to self-reporting by elected 
officials. In some processes, like budgeting, the survey shows is a tendency among treatment 
Upazilas to consult more widely. In other processes, like the measures to increase OSR the 
result is not clear; indeed the control Upazilas showed better performance than treatment 
LGUs. The last SDLG quarterly reports for 2013 also recognize that it will take some time to 
see major differences in Upazila governance and management as a result of SDLG interventions. 
1b. Unexpected Outcomes: The evaluation team did not find any outcomes unexpected by 
SDLG interventions or the PNGOs. 
1c. Added Value for Local Governance Sector: In developing its combination of helping 
LGUs increase local revenue and creating participatory input through the CiG mechanism, 
SDLG did provide added value to the LGUs themselves and to donor understanding of local 
governance. Its main governance innovation was the CiG understood as a “helping hand” for 
the local Council. It allowed them to increase their local OSR by strengthening the “outreach” 
of the elected Council members to their ward constituencies and better communicating the 
need for a local tax base.. The CiG groups also acted as “eyes and ears” for the Council, 
identifying and attending to critical issues like child marriage, domestic violence, trafficking and 
access to GOB safety net programs. 
Without taking anything away from this innovation, it still falls far short of the Task Order 
requirement to provide “comprehensive experimental programs to strengthen local 
government planning and management capacity… in quality and quantity of the compulsory duty 
services, as well as additional services, as appropriate, such as disaster mitigation and 
environmental conservation”. Nor were there any innovations generated with respect to the 
use by LGUs of “rigorous and practical indicators (and accompanying baseline measurements)…
to measure performance changes, such as surveys, tracking local revenue generation, etc.”,
which are also mentioned in the TO. 
The large number of partner LGUs covered by SDLG meant that the interventions were quite 
limited in terms of the LOE dedicated to capacity development in each partner LGU. With less 
than 10 days of formal training for each LGU, the project was severely constrained in both the 
depth of the capacity development, and in the number of functions/processes that it could 
cover in any LGU. Despite the TO requirements, it was unable to provide “demand driven” 
capacity development, for example to support, document and replicate innovative practices of 
individual LGUs outside of the three functional areas covered its standard package of training. 
This low level of capacity development support might explain some of the inconclusive results 
of the SDLG surveys. One question asked if there had been ward meetings; 41% of people in 
treatment LGUs said yes, as against 29% in control LGUs. However, when the paurashava 
residents were asked whether services had improved, there was no discernable difference 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 27
between treatment/control LGUs in basic urban services such as water, drainage, or solid waste 
management. While one interpretation might be that the perception of improvement will take 
time, another, more plausible explanation is that SDLG did not provide any “supply side” 
capacity development to the partner paurashavas improve those services. Indeed, this same 
point was mentioned by two of the paurashava Chairs interviewed by the evaluation team. 
2. Efficiency
With respect to efficiency in the project’s overall approach, in designing SDLG, USAID might 
have decided to focus project resources on either 1) a smaller set of local governments; 2) a 
single tier of local governments; or 3) LGUs within a single Division or a set of adjacent 
Districts. If the logic of SDLG was to create a “demonstration” of local government 
innovations of sufficient interest and impact to be taken up by the national government and 
other LGUs, this objective was made more difficult by the project approach. The dispersion of 
activities across a large number of LGUs led to a very “shallow” set of interventions in each of 
the participating LGUs, particularly in PS and Upazilas. 
It is not clear to the evaluation team why a group of 600 LGUs was a better “demonstration” 
than 300, 200 or even 100 LGUs. If there had been an effective link to the GOB for taking up 
innovations, a small number of LGUs would have sufficed. By focusing resources on a smaller 
sub-set of LGUs, within a single tier and/or a set of LGUs within a smaller geographical area, 
SDLG might have been able to free up resources for deeper interventions in improving 
governance and management, especially in the more complex LGU administrative structures of 
the paurashavas and Upazilas. 
3. Management and Administration
Implementing Partners: As specified in the TO, SDLG solicited and recruited nine 
domestic NGOs to implement Components 3 and 4, and SDLG assigned each organization a 
specific area to work in. The evaluation team interviewed staff from four of the nine PNGOs 
(one in each of the four districts visited). All four had considerable previous experience with 
field-based projects in their respective regions.The team met with the project coordinators 
(who managed overall SDLG implementation for their PNGOs) and the project officers (who 
dealt directly with the LGUs and CiGs) and learned about the training imparted by SDLG staff 
in a training-of-trainers (ToT) process that enabled the project officers to then train council 
members and CiG members, as well as to provide follow-on advice and guidance to them. 
The four PNGOs the team visited all had considerable experience implementing donor-assisted 
projects in different sectors. They appeared to adapt to SDLG with no difficulty. The project 
officers were impressive, committed to their SDLG work and to the LGU and CiG members 
they worked with. The best judges of their performance were the LGU chairs and councilors, 
and the CiG members, who uniformly expressed high satisfaction with the quality of their 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 28
training and the follow-up visits.24 The only negative comment the evaluation team heard was 
the frequently expressed complaint that SDLG’s activities had ended too soon and should 
continue longer to achieve results. 
One, related point about the PNGOs was their role in sustaining the CiGs. The interviews 
with CiG members and LGU officials coincided in highlighting the continuing role of the PNGO 
staff in convening and organizing the activities supported by SDLG, even in the third year of the 
project. In all but one of the UPs, the informants reported that the CiG meetings in the wards 
were organized in this way: PNGO staff contact the CiG leaders by telephone, agree on 
meeting logistics, and then call the members. The PNGOs often provided refreshments. 
Other Donors: It was not possible to ascertain if links were in place between other donor 
programs and SDLG at the field level among the sites the team visited. With respect to SDLG 
linkages with donor projects, the evaluation team met with the World Bank and the LGSP 
project management. While the TO requires SDLG to “strengthen and complement” LGSP, 
this did not happen in practice; there was no substantive collaboration. According to LGSP 
management and LGD officials, the linkages were limited to occasional “sharing meetings”. 
SDLG did not take up any of the activities suggested in the TO such as working with UPs 
receiving the extended block grant under the LGSP project “to pilot new implementation 
techniques”. For their part, the UNDP and ADB also reported that they had no collaboration 
with SDLG in their respective projects with UPs, Upazilas and paurashavas. 
Other USAID Sector Programs: The Task Order was explicit in the requirement for 
SDLG to explore opportunities with other USAID programs in health, environment, 
agriculture, disaster prevention or and food security. However, for the period covered by this 
evaluation, the evaluation team found that SDLG had made no progress in this respect. 
Locally Elected Officials: With only one exception among all the UP chairs, PS mayors, and 
elected council members at both levels, locally elected officials uniformly found SDLG’s training 
and CiG groups both helpful, as recounted elsewhere in this report. Thus, SDLG was perceived 
as being effective. Unfortunately, given the Upazila elections, it was not possible to reach a 
sufficient number of Chairs or Vice-Chairs to analyze their views on SDLG effectiveness. 
PNGO project officers reported experiencing considerable initial difficulty in presenting and 
selling the SDLG project to elected LGU councils. The selling point proved to be SDLG’s 
revenue generating potential. The project officers found their audience eagerly receptive to 
both the training they offered about local revenue generation and the idea of a CiG group that 
would assist in increasing the revenue to be collected. In effect, the councils were willing to 
accept the CiG concept and the risk of rivalry in order to increase local revenues. However, 
the councils soon found the CiG groups to be immensely in motivating participation in 
planning/budgeting; in identifying eligible recipients for social safety net programs; in mobilizing 
 
24
 Follow-up visits have often been a shoRCToming of field-based projects, but that was not the case here. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 29
people to participate in the ward shava and open budget meetings; and in acting as information 
gatherers about neighborhood needs such as clogged drains, broken streetlights, etc. 
Citizens: Apart from CiG group members, the evaluation team did not interview citizens 
directly on field visits. But the three opinion surveys sponsored by SDLG paint a picture of how 
citizens received and perceived the program. Details are reported above in subsection 1a. 
Government of Bangladesh: The Task Order states that SDLG will “build upon and deepen 
the existing relationship with the Local Government Division”. The specific requirements for 
Components 3 and 4 set out as expected results, “increased information available on 
participatory planning and community engagement to key actors including GOB officials at the 
local and national levels” as well as its “adoption and use…in local governments outside the 
purview of the SDLG program”. And “lessons learned … should be channeled into the national 
policy reform dialogue”. From interviews with SDLG management and GOB officials, and 
through a review of program reports, the team finds that SDLG achieved little or no impact in 
this respect. The senior officials in LGD were not aware of SDLG activities and reported 
having met only occasionally with project management to share information about activities. 
Overall, the evaluation team found none of the engagement with the GOB required by the TO. 
No GOB agency was involved in the planning of SDLG; the GOB was not consulted on 
selection of LGUs; there was no substantive collaboration with the GOB’s flagship program for 
UPs (LGSP); the GOB training institute, NILG, had no role in the design, testing or application 
of training materials; there was no collaboration with the UNDP/GOB Upazila Governance 
Program; and SDLG did not collaborate with the ADB/GOB UGIIP focused on paurashavas. 
4. Sustainability
Project Activities: The team’s timing was good for looking into sustainability of the 
interventions in local government units as the formal SDLG project had ended in December 
2013, two months before this evaluation. This timing allowed the team to check on post￾project interest in continuing the project’s work. In all of the locations visited, the team found 
uniform enthusiasm for continuing the CiG groups’ activities. The chairs/mayors, council 
members, and secretaries appreciated the increased revenue and the neighborhood information 
generated by the CiG groups. CiG members themselves enjoyed the enhanced self-esteem and 
local status the project gave them, as well as their role in selecting visible local improvements 
funded by the LGUs.25 Aside from the enthusiasm, however, the team did not find much 
evidence that the CiG groups had developed a capacity to carry on in making and keeping 
schedules, crafting agendas, organizing meetings, and providing logistical support. 
In their defense, it must be said that the CiGs actually had little time to build these skills. 
Though it formally launched in December 2010, SDLG had to recruit staff, locate offices, solicit 
and vet PNGO proposals, get the PNGOs approved by the GOB’s NGO Affairs Bureau, and 
 
25 The Insight study funded by SDLG (Unitrend, 2012) documented a quite similar finding among the CiG members. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 30
train the project officer trainers. The project officers then had to train the LGU members and 
set up and train the CiG groups. By the time the project officers completed this necessary 
work, it was summer and early fall 2012. 
The result was that only 12 to 18 months remained after all the training had been completed 
for field implementation with the CiGs in place working with the LGUs before project end in 
December 2013. Thus, the annual participatory planning/budget cycle and the OSR effort could 
only run for one or in some cases one-and-a-half rounds, and the neighborhood information 
generation was similarly restricted. This was enough time to build up significant momentum, as 
is clearly reflected in the keenness and even passion so widely expressed for the project, but 
likely not enough to enable it to continue on its own. In fact, as we discussed above, the 
evaluation team found in the field interviews with CiG members and LGU officials that given the 
short period for institutionalization of the new processes created with the CiG groups, the 
PNGOs still had a crucial role in organizing and sustaining CiG activities, convening and 
organizing many CiG activities. 
Measures to Enhance Sustainability: As we have discussed above, the evaluation team 
found that SDLG did not develop an ongoing relationship with GOB institutions like the LGD 
or NILG that would have remained in place after the project ended, to continue to provide 
LGUs with quality capacity development training and technical assistance. SDLG management 
has stated in the comments on the first draft of this evaluation report, that “SDLG’s primary 
sustainability objective was sustainability of results in the field”, with the idea that the councils 
and communities would continue the practices. As was discussed above, it remains to be seen if 
this has happened in practice; it would require another evaluation survey to assess SDLG’s 
results in this regard. 
SDLG’s explanation of “sustainability of field results” is not entirely consistent with the Task 
Order’s depiction of SDLG as a “demonstration project”, which clearly required the project to 
carry out specific activities to ensure a connection with GOB policy, and complementarities 
with other large donor programs that would sustain the SDLG innovations and extend them to 
other LGUs throughout the country. The evaluation team has found that these activities were 
not implemented, and as a result, this sustainability strategy through national-level linkages is 
not to be found in SDLG’s program interventions in Components 3 and 4.
Related to this same point is the absence of linkages between the innovations generated in the 
program Components 3 and 4 and the policy advocacy activities carried out under Components 
1 and 2. In the absence of GOB buy-in to the SDLG project, it did not have a ready audience 
for its governance and management results—as one senior LGD Secretary observed about 
SDLG, “it’s an NGO project”.
5. Relevance
For Components 3 and 4, there is no question that supporting local governance in general and 
SDLG’s particular approach to it will have continuing relevance in Bangladesh. As explained 
above in the discussion of “value-added”, SDLG’s interventions produced an important 
innovation in governance, through the CiG groups, which as a result led to a clear improvement 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 31
in LGU performance (in this case, increased tax revenues of UPs and paurashavas). Other 
SDLG interventions like participatory budgeting, or supporting service monitoring committees, 
while not innovative in the Bangladeshi context, are relevant for improving local democratic 
governance. 
With respect to Upazilas specifically, the evaluation team questions looked at whether SDLG’s 
interventions were relevant to the current situation of Upazila Councils in the context of the 
recent Upazila law. An analysis of the key sections of the law and a review of recent studies on 
local government in Bangladesh, including SDLG’s own policy documents, shows that the law 
sets up a conflict between the elected Council and the multitude of line ministries and agencies 
represented in the Upazila administration. Unlike decentralization policy in other countries, the 
Upazila law did not provide for internal reorganization or “reengineering” of processes within 
the line ministries and agencies. The Upazila continues to be just a physical infrastructure 
housing numerous de-concentrated GOB agencies, with a loose supervisory function carried 
out by the UNO. 
SDLG chose to train the Upazila Councils on three topics: service monitoring; planning; and 
financial management, with an emphasis on own source revenue, which are all relevant in 
principle. With respect to Upazila governance and management, SDLG might have considered 
how to make better use of the large amounts of information collected in health, education, 
social welfare, reproductive health, and agriculture, for Upazila oversight, benchmarking of 
services, etc. This would have been a very relevant intervention for the current Upazila legal 
framework, giving the elected Councils a tool to press for greater political accountability of the 
Upazila administration. Admittedly, it would have required greater buy-in from the GOB and 
focusing budget resources in fewer LGUs to support in-depth capacity development 
interventions. 
One notable absence in SDLG’s work with LGUs was improvement of information systems for 
decision-making, which was highlighted in the TO. In this sense, the evaluation team considers 
that SDLG lost an opportunity to respond to growing interest within the GOB generally in 
finding IT solutions to improve management. While it was not necessary to hinge the entire 
project on information systems, for example, SDLG might have explored opportunities with the 
PMO’s Access to Information (A2I) program, which is developing web portals for Upazilas. This 
would have been in line with the TO requirement to “help identify complementary areas of 
training and longer-term reforms or innovations that would bring further improvements”.
FINDINGS FOR COMPONENT 5
Component 5, officially titled “Windows of Opportunity” but often referred to as “Cross￾Cutting Issues” in SDLG’s reports, focused on women and youth. SDLG generally treated this 
as an add-on to the four main project components, as the term “windows of opportunity” 
implies and as is evident in the evaluation team’s SOW, where Component 5 is not listed as 
such along with the other components. Cross-cutting issues received little attention in the PMP 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 32
reporting.26 Even so, significant attention was devoted to women at both national and local 
levels. For instance, women UZP vice chairs received specific training as did MAB and BUPF 
national women’s committees. At the local level, all CiG groups were required to include 
women as one-third of their executive committee members, a goal that SDLG met. The project 
also trained women LG council members and in particular standing committee members in the 
UPs and PSs. There was no target for youth inclusion, but SDLG emphasized recruiting younger 
members, such that almost 12 percent of the committees consisted of citizens less than 25 
years old.27 In addition, SDLG sponsored the creation of several videos and radio programs 
featuring women and youth. 
The interview team interviewed 150 people in the site visits: 120 men and 130 women. The 
team was able to meet with women LGU and CiG members separately from their male 
counterparts at each LGU site visited, and a number of the women interviewed mentioned (in 
all cases with significant enthusiasm) the training they had received. Generally, they were more 
articulate than the male Council and CiG group members in explaining the objectives and 
contents of training. As for youth, while the team did not specifically ask to meet with younger 
CiG members, in most cases one or two members under 30 (which seemed the more common 
denominator of “youth” where the team visited) showed up to talk and were as enthusiastic 
about the project as their elders. 
CONCLUSIONS FOR COMPONENTS 3 AND 4 
1. Citizen Voice for Improving LGU Governance 
SDLG’s Major Achievement: The decision to focus on strengthening citizen voice through 
the CiGs, and to use the CiGs to support the LGUs in their OSR strategies was SDLG’s most 
outstanding achievement and a noteworthy innovation in the field of local governance in 
Bangladesh. As noted above, initially the revenue generation incentive was sufficiently attractive 
to lead LGUs to accept what they perceived initially as a threat of increased citizen involvement 
in the council’s business, and within quite a short timeframe, elected local officials came to see 
that the CiGs were useful in extending the reach of Councils to their constituents in the wards. 
During the project implementation period SDLG’s participating LGUs (UPs and PS) experienced 
higher growth of OSR than the control LGUs, which can be attributed to the campaigns led by 
Councilors and CiG members, and to SDLG training of officials on strategies for enhancing 
OSR. As noted above, after the project’s preparatory phases were put into place, only 18 
months remained in LOP for actual field implementation, i.e., just one or one-and-a-half budget 
cycles. In spite of this, it was sufficient to raise OSR dramatically, as is clear from the data for 
PMP 3.4 (Annex VI), which show a much larger increase for the treatment sample than for its 
control counterpart. 
 
26 Only one PMP Performance Indicator refers to women (1.4 Number of advocacy actions conducted by women 
elected representatives in local government), and none to youth. 
27
 These figures were reported by the PNGOs. The proportion of youth among CiG groups varied between seven 
percent and 17 percent among the nine PNGOs. The team had no means of verifying these numbers.
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 33
2. SDLG’s Use of Randomized Control Trials
On aspect of SDLG which should be highlighted is its use of Randomized Control Trails (RCT) 
approaches in the design of its PMP, which included a baseline measurement and two-follow –
up surveys over the life of the project. Much of what we can say about the effectiveness of the 
project is owed to the data collected through surveys over the life of the project. In this sense, 
SDLG is to be commended for its rigorous use of RCT methods in designing and implementing 
its PMP. Unfortunately, the team only had access to the data from the final survey during the 
report writing stage and was unable to look at the data in depth. However, it is certainly a rich 
source of information for USAID and should be utilized in subsequent project design. On the 
down side, the short project length meant that SDLG could not fully exploit this resource. 
3. Consequences of a Short Project Life 
While there were discernable impacts in own source revenue, SDLG did not show a clear 
“governance payoff”, expressed in greater trust in the treatment LGUs, or greater satisfaction 
with services. Indeed, citizen perception of improvement in actual service delivery shows little 
difference between treatment and control for the UPs (Table 7 in Annex V, also PMP 3.5a in 
Annex VI). The data also show a greater rise for the control than the treatment sample for the 
PSs (Table 8 in Annex V, also PMP 3.5b in Annex VI). The political culture question in the 
surveys, on citizen voice in ward shava meetings (PMP 4.5) produced muddled results; there 
was no indication that the treatment sample had made more progress than the control. 
These disappointing results on the SDLG governance indicators do not necessarily indicate 
project failure, but rather that the timeframe was too short to create and nurture a “demand 
side” CiG mechanism that could promote and sustain serious change in local political culture. It 
was in effect a misguided approach to the RCT concept to think that such a change could come 
so quickly and be expected to show up in opinion surveys conducted at short intervals. 
4. Low Level of Capacity Development Support 
Another explanation for the ambiguous survey results with respect to improvements in 
governance between the SDLG treatment and control LGUs might also be explained by the low 
levels of capacity development. SDLG’s interventions consisted of three short training courses 
on service monitoring, planning/budgeting, and financial management/OSR), with follow up 
facilitation by PNGO staff to motivate CiG members and Council members to implement the 
proposed policy/process changes. However, there was little or no attention on the “supply 
side”, i.e., the development of capacity within the local government administration itself. While 
this is understandable in the case of the UPs, with their scanty administrative structure, it is less 
so with respect to paurashavas and Upazilas, each with a complex administration. 
5. Sustainability of SDLG Field Programs
While the CiG involvement in the LGUs had gotten off to an excellent start and the evaluation 
team observed enthusiasm for carrying on the project from both LGU officials and CiG 
members, the team did not see evidence that CiG activities had become institutionalized to the 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 34
extent necessary to continue without outside TA and logistical support. The LGU officials and 
CiG members reported that the NGOs still had a major role in convening and organizing CiG 
meetings and activities. The cause was SDLG’s short LOP, which had not had time to build 
momentum within the CiGs for them to continue on their own. 
Without a follow-up survey after the end of the project, it is not possible to determine the 
degree of sustainability of the SDLG interventions. As was mentioned above, the short project 
time frame made it impossible to examine the long-term impacts of these interventions, 
especially with respect to changes in political attitudes and behavior. Thus, in such a short 
project period, the Randomized Control and Treatment (RCT) approach to impact evaluation 
used by SDLG to generate data for its PMP indicators was not able to achieve its full potential. 
The evaluation team agrees with SDLG’s comment on the first draft of this report, in the sense 
that “sustainability of field results can be determined objectively through a survey”.
6. PNGOs as Implementers 
Related to sustainability, is the question of SDLG’s use of NGOs as implementing partners in 
Components 3 and 4. The four PNGOs interviewed appeared to be effective implementers for 
SDLG, a judgment based on what the evaluation team learned from PNGO project officers and, 
more importantly, from the LGUs and CiG groups that were the recipients of training and 
follow-on guidance, and the achievements registered by treatment LGUs in terms of revenue 
generation. However, the decision to use PNGOs had negative implications for sustainability of 
the SDLG program interventions. 
This is in no way a criticism of the partner NGOs, who appeared to have performed well for 
SDLG. Rather, in practical terms, the expertise and capacity of NGOs developed under SDLG 
is lost to LGUs once the project is ended. In the absence of a private market for technical 
services to LGUs, in countries like Bangladesh governments play this role, through training 
institutes and other programs, sometimes led by line ministries (LGD) or training institutes 
(NILG). Granted, they have capacity issues; for this reason, USAID Forward objectives include 
developing sustainable institutional capacity and strengthening national systems. 
7. Exiguous Engagement with GOB 
The SDLG Task Order was explicit in spelling out USAID’s intention that the activities under 
this “demonstration project” would be taken up into GOB policy and widely replicated among a 
larger group of LGUs, saying “the Contractor shall maintain close contact with…GOB 
institutions, including the Local Government Division (LGD) throughout the life of this program 
to ensure cooperation and buy in from the GOB”. The evaluation team concludes that this did 
not happen. In the responses to the first draft of this evaluation report, SDLG noted that “in 
the third year and onwards” it started to interact with LGD, however, in practical terms, as it 
comes at the end of the project, it is difficult to assess its potential contribution to SDLG’s 
overall objectives. 
As a result, there has been no uptake into policy or wider replication within non-project LGUs 
of SDLG’s successes in effectively engaging communities in local governance and developing 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 35
capacities of community members and Councils to make local government more responsive. 
The evaluation team considers this to be a major, missed opportunity for USAID/Bangladesh. 
6. LESSONS LEARNED 
In this section, the team has tried to distill from its many findings and conclusions what it 
believes are the essential lessons to be learned from SDLG’s experience.
 Supporting research and advocacy for implementation of new LG laws is a worthy 
objective, but it requires deep and sustained engagement with GOB ministries and 
departments. 
 Short-term local governance programs may deliver impressive performance in LGU 
management processes in a brief project, but the sustainability of broader governance 
practices might require longer projects, or several projects working in the same group 
of LGUs over time. 
 The use of RCT approaches to measure impacts in local government projects is 
promising, but they need to be done in the context of longer project periods, or 
successive project interventions in the same LGUs. 
 In managing budget cuts to ongoing LG projects IPs should take care not to eliminate 
entirely those activities that will channel “demonstrations” and “models” into policy. 
Otherwise, the project’s broader development objectives could well be lost, despite 
successes in the field. 
 Working in LGUs with more complex administrative structures like paurashavas and 
Upazilas may require more attention to “supply side” governance improvements 
involving management and staff. 
 Local government projects that intend to have broader impacts should incorporate 
strategies for early and deep engagement with government institutions, involving them in 
project decisions. Otherwise the program interventions could well stay at the 
“demonstration” stage. 
 
RECOMMENDATIONS 
RECOMMENDATIONS FOR FUTURE PROGRAMMING 
1. Concentrate Resources 
The evaluation team recommends that USAID rethink the design of subsequent local 
government projects. To be able to influence policy change and reform, the team advises not 
spreading resources across all levels of local government. Rather, USAID should prioritize, 
focusing on either rural LGUs (UPs or Upazilas) or urban LGUs (paurashavas), according to its 
overall project orientation. Also, the project should consider carefully the number of LGUs 
required for a “demonstration”. A workable RCT approach could use as fewer treatment 
LGUs. 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 36
2. Ensure Continuity of Interventions 
For any effort promoting change in political culture to build up sufficient inertial momentum to 
sustain impact after EOP, the project must last longer than the 39 months of SDLG’s lifetime. A 
five- or six-year effort would be much better, but even that would in all likelihood represent 
just a beginning. 
Such an approach would take some re-jiggering of the 4- or 5-year LOP system, but it could be 
made to happen, perhaps with successor projects in which the RFP would require working in 
the same LGUs as the previous project. The point is to ensure continuity in the same LGUs; 
little is gained over time if each new project implementer moves on to a new set of partner 
LGs. In other words, a subsequent RFP should require the project to work in a sub-set of the 
existing SDLG 600 partner LGUs, whether UPs, paurashavas or Upazilas, chosen for their 
success in SDLG and/or through a field assessment to identify innovative leaders among the 600 
SDLG LGUs. 
3. Build in a Service Sector Focus 
Given USAID’s priorities in health, food security, WASH, and agriculture, the evaluation team 
recommends that the next local governance project focus on decentralizing service delivery in 
one or more of these sectors. In line with the other recommendations presented here, it 
should focus on a smaller set of LGUs, perhaps with a narrower geographical focus, to ensure 
that sufficient resources are available for intensive and sustained capacity development for 
management improvements, and to promote more “contagion” with adjacent LGUs. It should 
combine support for citizen participation and accountability mechanisms to strengthen both the 
“supply and the “demand” sides of local governance. In emphasizing responsiveness in service 
delivery, it should focus on LGUs with significant service functions (paurashavas and Upazilas) 
rather than UPs. 
4. Incorporate Change Management Interventions 
The governance focus of the local government project should be expanded to include piloting 
of “supply side” management solutions. Work in Upazilas could incorporate change manage￾ment and process reengineering approaches, working with line ministries to experiment with 
governance improvements. Returning to the previous recommendation, it should be noted that 
this is intensive in training, facilitation, mentoring, etc., well beyond the 10 days of training 
provided by SDLG. Experience from decentralization programs in other countries suggests that 
it could require a full time trainer/facilitator in each Upazila or paurashava to lead the change 
management process, supported by experts in process reengineering, and in close collaboration 
with GOB counterparts. 
5. Look at IT Solutions 
Related to the previous point, in designing the RFP, USAID should look carefully at how UPs 
and Upazilas can improve management/governance between levels, particularly through better 
use of information on plans, budgets, and service provision levels and standards. What is 
notable about the SDSLG project in this aspect is its lack of attention to IT solutions, which 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 37
have a huge potential for bringing local governments into a performance management approach, 
and also incorporating citizen access and oversight. This might involve piloting innovative IT 
solutions (e.g. a simple information management systems that can piggy back on the existing 
Union Information Service Centers), or better access to existing data sources (i.e., service data 
collected by Upazila offices of ministries and agencies) and its use for decision-making and 
accountability. 
6. Engage with GOB in Project Implementation 
The history of local government reform—indeed, any policy reform—in Bangladesh suggests 
that without taking the bureaucracy on board it is virtually impossible to implement policy 
change. The discussion of “management solutions” in a previous recommendation suggests, too, 
that future local government projects need to dedicate more effort to engagement with the 
GOB, recognizing that government agencies have increasing fiscal, human and technical 
resources—indeed, LGD’s own budget has grown by almost 10% annually in real terms since 
2009. The sustainability of project interventions will best served through direct involvement 
with the GOB, in both project design and implementation, to ensure buy-in from relevant 
ministries and agencies. 
If a future local government projects, it is recommendable to work with LGD and NILG in the 
design of the project, and in the implementation of capacity development and training. Evidently, 
given the legitimate concerns about NILG’s technical capacity, the project would have to 
include a capacity development component for NILG itself, perhaps drawing on existing NGO 
expertise and staff for assistance in design, in training of trainers and in monitoring training 
quality. 
If the project includes Upazilas, it should also engage line ministries and agencies responsible for 
service delivery, to lead, design and implement the required change management strategies for 
Upazila governance, perhaps even the PMO’s Governance Innovations Unit. This would require 
prior engagement and negotiation between USAID, the PMO and the respective ministries, 
before issuance of an RFP. The Task Order should set out in detail the project counterparts in 
line ministries, LGD, etc., the specific service/processes that will be reengineered, and the 
modalities for GOB participation, which might include colocation in ministries and/or Upazila 
complexes. 
Local government projects are particularly conducive to the use of benchmarking governance 
and management improvements, comparing LGU performance over time. In engaging LGU and 
GOB actors, subsequent projects could incorporate governance and management indicators, 
developed collaboratively with the LGU leaders and in the case of Upazilas, with the GOB 
(LGD and line ministries). 
7. Engage/Coordinate/Collaborate More with Other Donors 
The Task Order requires the Contractor to coordinate closely with other donors, and 
specifically LGSP. Our findings show that this did not happen. The team suggests that USAID 
Final Performance Evaluation of the Strengthening Democratic Local Governance (SDLG) Project 38
could profitably explore how to complement and extend the impacts of other donor programs 
in the field of local government, such as: 
 For union parishads, the World Bank’s LGSP and the Swiss Development Corporation.
 For paurashavas, the Asian Development Bank’s UGiiP, JICA, GiZ.
 For upazila parishads, the UNDP and A2I in the PMO. 
Whatever the focus of any future project, effective engagement with other donors (and with 
the GOB) would require more upfront work by USAID prior to issuing the RFP/RFA. The 
team recommends that USAID undertake a detailed assessment to identify opportunities for 
cooperation with other projects, and negotiate with the respective donors and the GOB to 
specify the exact modalities and contents of cooperation with USAID’s selected implementing 
partner. These details should then be incorporated into the RFP/RFA and the respective Task 
Order. 
 
8. Separate Contract Mechanisms for LGA Support 
USAID might consider leaving out LGA support/advocacy from a future project, and look 
instead for a separate contract mechanism under USAID Forward. These might include direct 
grants, conditional on pre-grant enquiries and appropriate capacity development prior to 
approval of funding. Such a grant should maintain the established performance criteria of 
internal democracy through elections of the national and regional executives, and a minimum 
level of membership and fees collection to sustain the association. Such a grant might be 
implemented in cooperation with another international organization (Commonwealth Local 
Government Forum), or an association of local governments from a neighboring country (the 
Philippines or India). 
9. Continue to Use Rigorous Evaluation Approaches 
The evaluation team recommends that USAID continue to use RCT approaches to evaluate the 
impacts of its interventions in strengthening local government, with a long-term focus on how 
these interventions change not just management processes, but also governance variables, 
including people’s knowledge of local politics, attitudes towards local governments, perceptions 
of service delivery, and patterns of political engagement and participation. In particular, the 
evaluation team recommends that USAID support another application of SDLG’s survey of 
treatment and control LGUs, in 2015 or 2016, to examine the sustainability and mutation of 
project interventions. The results of such a study should be invaluable to USAID/Washington 
and Missions worldwide in providing a rare picture of what contributes to sustainability in its 
local governance work. 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 39
ANNEXES 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 40
ANNEX I: EVALUATION SCOPE 
OF WORK 
Scope of Work 
for the Strengthening Democratic Local Governance (SDLG) 
External Performance Evaluation 
USAID/Bangladesh 
Office of Democracy and Governance 
Program Identification Data 
Program Title : Strengthening Democratic Local Governance (SDLG) 
Program Number: EPP-I-00-04-00035-00, Task Order No: AID-388-TO-11-00001 
Program Dates : Start Date: 12/21/2010- End Date: 3/30/2014
Program Funding: $ 19,248,165 
Implementing Organization: Tetra Tec ARD 
Contracting Officer Representative (COR): Sherina Tabassum 
I. Background 
While the constitution of Bangladesh calls for a strong locally-elected government at each administrative 
level, what primarily exists is a weak form of sub-national government, structured in a way that allows it 
to play a minimal role in development. According to national law, each administrative unit of local 
governance (Union Parishad, Upazilla Parishad, municipalities, etc.) was to consist of elected 
representatives in charge of public offices and services. In addition, these local bodies were to have 
taxation authorities. The Union Parishad (UP) level is the one level that has consistently had elected 
representatives. And, it was only in January 2009, based on the legal framework established by the 
Caretaker Government (CTG) that governed Bangladesh from January 2007 – December 2008 that 
three leadership positions at the Upazilla-Parishad level were directly elected for the first time. 
In most levels of local government, little potential exists for generating revenue from permitted sources. 
The majority of resources from the central government flow to the various line ministries to programs 
managed by the line ministry staff, over which local government has minimal influence. In addition, at 
various times, local government structure changes depending on who is in power. The end result 
provides few opportunities for Bangladeshis to influence funding and decision-making at the local levels. 
Despite these challenges, there has historically been a vigorous national debate about what should be 
the main level of elected local government and their related authorities. Many rural Bangladeshis view 
the Upazilla Parishads level as closest to the people, and it is generally regarded more favorably than 
other elected and posted government officials. 
As part of the wave of reforms pushed by the CTG, several ordinances focused on establishment of a 
more robust legal framework for local government that provided for increased authorities, particularly 
for Upazilla Parishads. Specifically, it made the chairman and two vice chairman directly elected by 
voters, increased funds they would manage, and incrementally changed the Upazilla relationship to line 
ministry officials posted at the Upazilla level. A setback to these reforms occurred in April 2009, when 
the new administration enacted a law that failed to provide the newly elected Upazila officials with many 
of the authorities outlined in the CTG ordinance. Now according to the current law, Upazila elected 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 41
officials must seek advice from the Member of Parliament of that locale in decision-making. In practice, 
this undermines the Upazila councilors who are also directly elected by the people like MPs. 
The response to the new law among locally elected representatives and key constituencies including the 
two national local government associations, the Bangladesh UP Forum (BUPF) and Municipal Association 
of Bangladesh (MAB), was fast and vigorous. A heated debate has been underway, and in October 2009, 
a writ petition was filed by the BUPF to Bangladesh’s High Court to deem the law unconstitutional. 
Through USAID’s assistance, both BUPF and MAB have made significant progress in building a national 
constituency for local governance reform. The associations now have elected Executive Committees, a 
secretariat, and their own staff. The BUPF includes about 1,000 paid members, while the MAB has 
brought all 309 municipalities under its fold. These associations have also begun to provide a platform 
for local governments to advocate for greater autonomy in local decision-making and for a larger share 
of the national budget. 
Women’s participation at the local level has increased substantially over the last 15 years. A major 
change occurred in the rural areas in the mid-1990s when one-third of the Union Parishad seats were 
reserved for women. Immediately, the number of elected women officials grew to approximately 
12,000. More recently, one of the three newly-elected positions to the Upazilla Parishad is reserved for 
a woman; however, the guidance from the Ministry of Local Government and Rural Development 
designated no authorities for these elected women. 
The USAID/Bangladesh Mission overall strategy on local government has focused on the promotion of a 
legal and policy framework conducive to effective, participatory local governance. First, USAID funded a 
program from 2001-2005 through ARD, Inc. entitled the National Constituency for Strong Local 
Government. This was followed by a two-pronged effort to support local governance starting in 2005 
through activities in 134 elected rural and urban governments implemented by RTI International and the 
Bangladeshi organization, Rupantar, through the Democratic Local Governance Program (DLGP) and 
the Improving Local Governance by Strengthening Union Parishad and Creating Citizens’ Awareness 
(ILLG) program, respectively. The DLGP concluded in 2008, but ILLG, which ended in March 2011, 
remained active and expanded its scope and worked with a total of 214 local governments (210 Upazilas 
and four Municipalities). Interventions under ILLG encouraged national dialogues on decentralization 
policies to increase access to more resources and autonomy for local governments. 
In April 2008, USAID conducted an internal (USAID/Washington) evaluation of the Democratic Local 
Governance program, which included the RTI and Rupantar programs. The evaluation sought to 
provide insight on what has worked and what has not and offer recommendations for the next phase of 
local government activities. The assessment findings and recommendations have greatly informed the 
design of the SDLG program. The most significant achievements cited in the report included: 1) 
improvements in service delivery as a result of training and increased interactions between Union 
Parishads and citizens that helped to establish community priorities; 2) an increase in revenue generation 
in 105 Union Parishads by over 50 percent; 3) establishment of 85 (at the time of the report) Citizens 
Forums that enabled citizens for the first time to participate in local decision-making in a structured and 
constructive way; and 4) the impact and growing capacity of the BUPF and MAB to generate 
support/interest and advocate for local governance reform. 
Some of the weaknesses cited included the need to improve measurements on the quality and quantity 
of service delivery as well as the need for improved coordination with other donors and within 
USAID/Bangladesh on local governance issues and programming. Donor coordination is especially 
important as USAID has and will, through this SDLG program, work with a relatively small subset of 
local governments. Therefore, USAID’s main comparative advantage is in providing demonstration sites 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 42
or models for improved service delivery and citizen participation, in which the creation and 
dissemination of best practices is paramount. In addition, USAID efforts to provide support to the local 
government associations will continue to be instrumental in building a national constituency for reform 
and advocacy. 
It is important that the SDLG program not only build on the USAID/Bangladesh Mission’s past and 
current local government strengthening programs, but also seek synergies with other USAID and donor 
programs. The program is also designed to take into account the salient aspects of the 
USAID/Bangladesh Mission programs implemented under different Objectives. 
SDLG Result Framework 
Component 1 – Roles and Authorities of Local Governments 
Policy reform for greater roles and authorities for local governments depends on the will and 
commitment of policy makers. Constructive policy debate and sharing of experiences are critical to 
increase policy makers’ understanding of local government issues. In addition, sub-national governments 
in Bangladesh can and should participate more actively in this debate on the development of policies and 
legal reforms that affects their roles and resources. With support from USAID under DLGP, the BUPF 
and MAB have been relentlessly advocating for policy reform. Local governments’ roles change not only 
as result of legislative and regulatory actions but also through the gradual expansion of pilot practices or 
innovations initiated in response to felt needs or citizen demands. For example, the Mission’s work 
along with other donors on providing greater fiscal autonomy to the Union Parishads resulted in 
allocation of small direct block grants under the annual development budget from the Local Government 
Ministry directly to the Union Parishads. Although the amount of that grant was less than one percent of 
the annual development budget, that was a major breakthrough in the traditional policy and practice of 
channeling funds to UPs through the district and sub-district administrations of the central government. 
Subsequently, with the loan support from the World Bank and grants from some other donors, the 
Ministry of Local Government increased the amount of extended block grant for UPs under the LGSP. 
Component 3 –
Transparent and Effective 
Service Delivery by Local 
Governments
Component 2 –
Advocacy and Capacity 
Building of Local 
Government Associations 
Component 1- Roles 
and Authorities of Local 
Governments
Strategic Objective: More effective and responsive democratic institutions and practices
Component 4 –
Citizen Participation 
in Local Decision 
Making
Component 5- Windows of Opportunity 
(crosscutting)
Program Goal: Improve transparency and participatory public administration at the sub-national level and 
enhance legal and policy reform at the national level in order to promote and expand decentralization 
Figure 1 RESULTS FRAMEWORK
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 43 
The objective of this component is to plan and carry out specific advocacy activities to encourage the 
national government to adopt legal and policy reforms to expand roles, authorities and resource 
allocation for local governments so that they can provide better service delivery. 
Component 2 – Advocacy and Capacity Building of Local Government Associations 
Capacity building is crucial for both the BUPF and the MAB to play an even greater role as the leading 
advocacy groups for local governments in the country. These two national level associations currently 
represent approximately one-third of all the local government units in Bangladesh and are poised to take 
the lead in promoting local governments’ concerns. Despite great success in advocating for reform and 
building a national constituency for local government, the capacity of the BUPF and MAB is mixed. 
Although the organizations do have a growing membership base, they are far from being financially self￾sustaining. Further, the associations continue to sort through challenging issues related to management, 
organization and general operations (legal, administrative, etc.). In addition, with elected Upazila 
Parishads in place, there may be a need for the formation of a new association that will represent the 
sub-national government at this level. Currently, the Upazila elected officials are raising their voice for 
policy change, but their advocacy is ad hoc. A forum of Upazila Parishads representing their collective 
voice could join the policy advocacy efforts carried out by BUPF and MAB. 
The objective of building capacity of the local government associations is to increase their ability to 
advocate and lobby on behalf of the sub-national governments that bring about broad policy changes that 
empower local governments and advance democratic decentralization. 
Component 3 – Transparent and Effective Service Delivery by Local Governments 
Successful decentralization depends on the ability of local governments to deliver effective public 
services in a transparent and efficient manner. To that end, local government elected officials and 
managers need effective management tools to expand and improve service delivery in cost-efficient 
ways. There is a wide array of potential in-country capacity that can be drawn on and/or mobilized, 
including research institutes, NGOs and the private sector. In addition, Information and 
Communications Technologies may offer cost-saving and effective approaches for improved 
performance of local government managers and administrators. 
The principal emphasis of this component is focused on achieving an immediate impact on local 
government management capacity through on-the-job training and technical assistance. The program 
should also help identify complementary areas of training and longer-term reforms or innovations that 
would bring further improvements. The role of the program is to generate new ideas and innovations 
through discussion, debate and consensus building. Lessons learned in this component should be 
channeled into the national policy reform dialogue. 
The objective of building the capacity of the targeted local governments is to enable them to be effective 
and transparent service providers at the sub-national level. 
D. Component 4 – Citizen Participation in Local-Decision Making 
Transparency in public management and in the use of public funds is fundamental to citizens’ trust and 
confidence in their local governments. Transparency can only be established by promoting and requiring 
the use of participatory planning by local governments and an active involvement of the citizenry in the 
affairs of the elected councils. USAID/Bangladesh Mission assessments and observations indicate that 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 44
initiating participatory processes is important to bridge the gap between elected officials of local 
governments and their constituents. 
The objective of increasing citizen participation in local decision-making is to ensure transparency and 
accountability in public management will lead to improved service delivery. 
Gender Perspective of the program: 
The SDLG activity supports women and men to enhance their leadership and management skills. 
Leadership training for women local representatives and women local council candidates is a priority for 
this project. Training for the newly elected women Vice Chairs of the Upazila Parishads as well as the 
Union Parishad representatives in their roles and responsibilities and other related issues is a major 
focus of this contract. 
The dynamics of women’s and men’s relationships as well as the complexities and inter-relatedness of 
their roles in improving governance at the local level cannot be simply addressed by any single activity. 
However, ensuring that women receive the leadership and management skills necessary to be successful 
at the local level is essential in promoting women as capable and reliable leaders. Likewise, they should 
be recruited and trained for local government jobs. The program promotes the concept of expanded 
roles and responsibilities for women through various seminars, workshops and conferences. 
Youth Perspective of the program: 
Bangladesh has a large youth population. There are approximately 45.7 million youth between the ages 
of 10 to 24, representing around 32% of the total population. Youth have enormous potential to play a 
pivotal role in promoting local level democracy by helping the local governments serve their 
constituents effectively. The program tries to involve the youth in the Citizens’ Forum, which will not 
only provide them with the opportunity to play a watchdog role but also help build them as future 
leaders. 
II. Objectives of the Evaluation 
The objective of the performance evaluation is to measure the development outcomes of the program 
with a view to drawing lessons learned for the selection, design, and implementation of future projects. 
The performance evaluation will also assess the relevance and sustainability of the program outcomes. 
The evaluation will: 
 Assess SDLG program’s actual results against targeted results;
 Assess the efficacy of the SDLG implementation tools and management structure in meeting the 
objectives; 
 Make recommendations to USAID/Bangladesh concerning future programming with Local 
Government. 
The audience for this evaluation is USAID/Bangladesh, USAID/Washington leaders of USAID Forward, 
other USAID missions, The Tetra Tec ARD, relevant stakeholders such as BUPF, MAB, Elected Local 
Government, Community leaders, community people and existing USAID implementing partners etc. 
III. Evaluation Questions 
The evaluation should review, analyze, and evaluate the SDLG program by answering the following 
evaluation questions, and where applicable, identify opportunities and make recommendations for future 
programming with Local Government. In answering these questions, the Evaluation Team should assess 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 45
both the performance of USAID and that of the implementing partner(s). The evaluation questions, in 
order of priority are: 
 
7. Effectiveness: To what extent has the SDLG program been successful in achieving its planned 
objectives? To what extent has the program resulted in unexpected outcomes (positive or 
negative)? The Local Governance sector in Bangladesh receives the largest amount of 
development assistance. What added value has USAID brought to the sector through SDLG? 
What has SDLG done differently? 
8. Sustainability: How sustainable are SDLG program activities, and what measures could have 
been taken to enhance sustainability? 
9. Efficiency: Are the objectives being achieved economically by the project intervention? 
(Comparison: resources applied – results)? Is there evidence from the implementation of SDLG 
to suggest that alternative program approaches may have been more successful? 
10. Relevance: To what extent are the project’s objectives still relevant to the current 
development circumstances in Bangladesh, and will they provide sufficient guidance for 
appropriate programmatic and technical assistance decisions? 
11. Management and Administration: How effective and flexible has the SDLG management 
been in working with implementing partners and beneficiaries, such as local government units, 
locally elected officials, citizens, and GOB? 
12. Cross-cutting Issues: To what extent were gender, youth, and disability issues addressed by 
SDLG’s interventions in the targeted areas? 
IV. Proposed Evaluation Methodology 
The detailed methodology of this evaluation will be described by the evaluation team in the Work Plan; 
this will include presentation of an evaluation matrix that will explicitly link evaluation questions and sub￾questions to particular data collection approaches and data sources. 
In general, the evaluation will apply a mixed-methods approach, with an emphasis on comparative field￾based case studies of Local Government Units. Some quantitative analyses may be featured, for 
example, in the review of SDLG’s performance monitoring data or in the analysis of the program’s 
efficiency. The qualitative side of the evaluation will be incorporated to address several questions 
(regarding program relevance, management and administration, and sustainability, for example). In 
addition, the field data collection will involve intensive case study visits, organized around a set of semi￾structured individual interviews and group discussions. Individual interviewees will include: members of 
Local Government bodies, staff from Local Government ministries, staff of Donor organizations working 
with Local Government, Local opinion leaders, general community people, etc. The team will welcome 
suggestions from USAID as well as The Tetra Tech ARD and other evaluation stakeholders, for 
additional data sources at the community level. Discussion groups will include balanced numbers of men 
and women; in addition, as appropriate to local circumstances sex- or age-segregated discussion groups 
will be used to promote free discussion by women, men, and youth. 
The evaluation team will analyze the information collected to establish credible answers to the questions 
and provide major trends and issues. USAID requires that evaluations explore issues of gender; thus, 
the evaluation should examine gender issues within the context of the evaluation of SDLG activities. 
Methodological limitations and challenges for this evaluation are expected to include: 
 Ensuring adequate representation of interview and rapid appraisal sources vis-à-vis the full 
scope of SDLG activities and outcomes; and 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 46
 Taking systematic actions to counter any biases in (a) reporting by data collection sources and 
(b) interpretations of collected data by the evaluation team. 
The methodology narrative should discuss the merits and limitations of the final evaluation methodology. 
The evaluation team will design appropriate tools for collecting data from various units of analysis. The 
tools will be shared with USAID during the evaluation and as part of the evaluation report. 
The evaluation team will be required to perform evaluation tasks in Dhaka, Bangladesh and also will 
travel to activity sites within the country 
VI. Existing Sources of Information 
USAID/Bangladesh DG Office will provide documents for the desk review that are not available outside. 
The list of available documents is presented in Annex A. The list is not exhaustive and the Evaluation 
Team will be responsible for identifying and reviewing additional materials relevant to the evaluation. 
The USAID/DG office will also help the evaluation team with contact information for relevant 
interviewees. 
VII. Deliverables 
All deliverables are internal to USAID and the evaluation team unless otherwise instructed by 
USAID. Evaluation deliverables include: 
Evaluation Team Planning Meeting (s) – essential in organizing the team’s efforts. During the 
meeting (s), the team should review and discuss the SOW in its entirety, clarify team members’ roles 
and responsibilities, work plan, develop data collection methods and instruments, review and clarify any 
logistical and administrative procedures for the assignment and prepare for the in-brief with 
USAID/Bangladesh; 
Work Plan - Detailed draft work plan (including task timeline, methodology outlining approach to be 
used in answering each evaluation question, team responsibilities, and data analysis plan): Within 5 
working days after commencement of the evaluation; 
In-brief Meeting - In-brief with USAID/Bangladesh: Within 2 working days of international team 
members’ arrival in Bangladesh;
Evaluation Design Matrix – A table that lists each evaluation question and the corresponding 
information sought, information sources, data collection sources, data analysis methods, and limitations. 
The matrix should be finalized and shared with USAID/Bangladesh before evaluation field work starts. It 
should also be included as an annex in the evaluation report. 
Data Collection Instruments – Development and submission of data collection instruments to 
USAID/Bangladesh during the design phase and after the evaluation is completed; 
Regular Updates - The Evaluation Team Leader (or his/her delegate) will brief the BDGPE COR on 
progress with the evaluation on a weekly basis, in person or by electronic communication. Any delays 
or complications must be quickly communicated to USAID/Bangladesh as early as possible to allow quick 
resolution and to minimize any disruptions to the evaluation. Emerging opportunities for the evaluation 
should also be discussed with USAID/Bangladesh. 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 47
Debriefing with USAID - Presentation of initial findings, conclusions and preliminary 
recommendations to USAID/Bangladesh before the international team members depart from 
Bangladesh. 
Debriefing with Partners - The team will present the major findings from the evaluation to USAID 
partners (as appropriate and as defined by USAID) through a PowerPoint presentation prior to the 
team’s departure from the country. The debriefing will include a discussion of achievements 
and activities only, with no recommendations for possible modifications to project approaches, 
results, or activities. The team will consider partner comments and incorporate them appropriately in 
drafting the evaluation report. 
Draft Evaluation Report - – The Evaluation team will analyze all data collected during the evaluation 
to prepare a draft Performance Evaluation Report and submit the report within 10 working days on 
after the departure of international team members from Bangladesh. The draft report must be of a high 
quality with well-constructed sentences, and no grammatical errors or typos. The report should answer 
ALL the evaluation questions and the structure of the report should make it clear how the evaluation 
questions were answered. The draft report must meet the criteria set forth under the final report 
section below. USAID will provide comments on the draft report within ten working days of submission. 
The Evaluation Team will in turn revise the draft report into a final Performance Evaluation Report, fully 
reflecting USAID comments and suggestions, within five working days of receipt of the written 
comments; 
Final Report: The Evaluation Team will submit a final Performance Evaluation Report that incorporates 
Mission comments and suggestions no later than five working days after USAID/Bangladesh provides 
written comments on the draft Performance Evaluation Report. The format of the final report is 
provided below. The report will be submitted in English, electronically. 
The final report should meet the following criteria to ensure the quality of the report: 
 The evaluation report should represent a thoughtful, well-researched and well organized effort 
to objectively evaluate what worked in the project, what did not and why. 
 Evaluation reports shall address all evaluation questions included in the scope of work.
 The evaluation report should include the scope of work as an annex. All modifications to the 
scope of work, whether in technical requirements, evaluation questions, evaluation team 
composition, methodology or timeline need to be agreed upon in writing by the technical 
officer. 
 Evaluation methodology shall be explained in detail and all tools used in conducting the 
evaluation such as questionnaires, checklists and discussion guides will be included in an Annex 
in the final report. 
 Evaluation findings will assess outcomes and impact on males and females. 
 Limitations to the evaluation shall be disclosed in the report, with particular attention to the 
limitations associated with the evaluation methodology (selection bias, recall bias, unobservable 
differences between comparator groups, etc.). 
 Evaluation findings should be presented as analyzed facts, evidence and data and not based on 
anecdotes, hearsay or the compilation of people’s opinions. Findings should be specific, concise 
and supported by strong quantitative or qualitative evidence. 
 Sources of information need to be properly identified and listed in an annex. 
 Recommendations need to be supported by a specific set of findings. 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 48
 Recommendations should be action-oriented, practical and specific, with defined responsibility 
for the action. 
The format of the final performance evaluation report should strike a balance between depth and length. 
The report will include a table of contents, table of figures (as appropriate), acronyms, executive 
summary, introduction, purpose of the evaluation, research design and methodology, findings, 
conclusions, lessons learned and recommendations. Where appropriate, the evaluation should utilize 
tables and graphs to link with data and other relevant information. The report should include, in the 
annex, any dissenting views by any team member or by USAID on any of the findings or 
recommendations. The report should not exceed 30 pages, excluding annexes. The report will be 
submitted in English, electronically. The report will be disseminated within USAID. A second version of 
this report excluding any potentially procurement-sensitive information will be submitted (also 
electronically, in English) to Development Experience Clearinghouse (DEC) for dissemination among 
implementing partners and stakeholders. 
All quantitative data, if gathered, should be (1) provided in an electronic file in easily readable format; (2) 
organized and fully documented for use by those not fully familiar with the project or the evaluation; (3) 
owned by USAID and made available to the public barring rare exceptions. A thumb drive with all the 
data could be provided to the COR. 
The final report will be edited/formatted by Social Impact and provided to USAID/Bangladesh 15 
working days after the Mission has reviewed the content and approved the final revised version of the 
report. 
VII. Team Composition/ Technical Qualifications and Experience Requirements for the 
Evaluation Team 
The evaluation team will include and balance several types of knowledge and experience related to 
program evaluation. Individual team members should have the technical qualifications as described 
below: 
1. Team Leader: An international Senior Evaluation Specialist (Program Development Specialist –
Senior) with experience in evaluating Local Government programs in developing countries. The 
Team leader will provide leadership for the Team, finalize the evaluation design, coordinate 
activities, arrange meetings, consolidate individual input from Team members, and coordinate the 
process of assembling the final findings and recommendations. S/he will also lead the preparation 
and presentation of the key evaluation findings and recommendations to USAID/Bangladesh. S/he 
should also have the number of years of experience and level of education required under the 
Program Development Specialist – Senior labor category. Experience in conducting assessments 
and designing strategic responses to Local Government in developing countries is required. 
Ability to produce highly quality evaluation report in English is essential. 
2. Evaluation Methodologist: At least 5 years of experience in designing and conducting field￾based evaluations and assessments in the democracy and governance sector. Experience in 
conducting assessments and designing strategic responses to Local Government in developing 
countries is required. Relevant experience in Bangladesh preferred. 
3. National Team Member: A national Senior Sector Specialist should have working experience 
with Local Government in Bangladesh. At least ten (10) years of experience in democracy and 
governance programs and some experience managing or implementing programs related to Local 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 49
Government in developing countries is required. Ability to conduct interviews and discussions 
and write well in English is essential. 
4. National Team Member: A national senior or mid-level evaluation specialist should have at 
least 7 years of experience in designing and conducting field-based evaluations and assessments in 
the democracy and governance sector. Relevant experience in Bangladesh preferred. 
The proposed team composition will include one team leader and two/three team members. USAID 
strongly encourages the team to have one member from the LTTA staff for this Evaluation. All positions 
will be considered key staff and will require USAID approval. 
Overall the team will need expertise in USAID practices and expectations in program evaluation; 
program design and analysis; quantitative and qualitative data collection and analysis; survey design and 
analysis; program issues, innovations and challenges in promotion of public sector transparency and 
accountability; and USAID practices and requirements in program performance measurement. 
VIII. Conflict of Interest 
All evaluation team members will provide a signed statement attesting to a lack of conflict of interest, or 
describing an existing conflict of interest relative to the project being evaluated. USAID/Bangladesh will 
provide the conflict of interest forms. 
IX. SCHEDULING AND LOGISTICS 
Work is to be carried out over a period beginning from mid-February 2013, with field work completed 
in March, 2014 and final report and close out concluding o/a May, 2014. 
Funding and Logistical Support 
The proposed evaluation will be funded and implemented through the BDGPE project. Social Impact 
will be responsible for all off-shore and in-country administrative and logistical support, including 
identification and fielding appropriate consultants. Social Impact support includes arranging and 
scheduling meetings, translation services, international and local travel, hotel bookings, working/office 
spaces, computers, printing, photocopying, arranging field visits, local travel, hotel, and appointments 
with stakeholders. 
The evaluation team should be able to make all logistic arrangements including the vehicle arrangements 
for travel within and outside Dhaka and should not expect any logistic support from the Mission. The 
team should also make their own arrangement on space for team meetings and equipment support for 
producing the report. 
Please note that business calendar days and LOE days do not match up, as the evaluation team will not 
be working full-time on this evaluation while in the U.S. Extending calendar time past just the number of 
days of LOE allows SI time to review the evaluation report for quality and to provide comments back to 
the team to be addressed. This extra time does not cost USAID extra money, it simply improves the 
quality of the documents submitted. 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 50
Task/ Deliverable Proposed 
Dates
Business 
Calendar 
Days*
LOE Days (estimated, days)
Team 
Leader
Evaluation 
Specialist National National
Review background documents & 
preparation work (offshore): Draft work 
plan submitted to SI by 2/14 and to 
USAID/Bangladesh by 2/21 (Dhaka time)
1/27-2/20 15 3 3 3 3
Travel to Bangladesh by expat team 
member 2/20-2/22 3 2 - -
Team Planning Meeting hosted by 
BDGPE 2/23 1 1 1 1 1
In-brief with USAID/Bangladesh 2/24 .5 .5 .5 .5 .5
Meet with SDLG/Tetra Tech ARD staff 2/24 .5 .5 .5 .5 .5
Data collection 2/25-3/15 16 16 16 16 16 Analysis and product drafting in-country
Final work plan due to USAID (draft due 
from team to SI by 2/23) 2/26 Counted 
above
Counted 
above
Counted 
above
Counted 
above
Counted 
above
Evaluation Team submits draft 
presentation to SI on 3/11 and then to the 
USAID/Bangladesh DG Team for review
on 3/12; data collection continues after 
submission
3/12 Counted 
above
Counted 
above
Counted 
above
Counted 
above
Counted 
above
USAID provides comments (as needed) 
on report outline and draft presentation 3/13 Counted 
above
Counted 
above
Counted 
above
Counted 
above
Counted 
above
Presentation and debrief with DG Team 
and USAID/Bangladesh 3/16 .5 .5 .5 .5 .5
Debrief meetings with key stakeholders, 
including GOB 3/16 .5 .5 .5 .5 .5
Expat Team members depart Bangladesh 3/17-3/18 2 2 - - -
Produce draft report. Team leader delivers 
draft report to SI/BDGPE on 15 Nov. 3/19-4/11 18
SI reviews draft report and provides 6 6 3 3
comments back to the team to address; 
delivers to USAID on 5/2
4/12-5/2 15
USAID and partners review draft and 
provide comments by 5/15 5/3-5/15 10 - - - -
Team revises draft report and submits to 
SI by 5/16; SI comments back to the team 
and then once final, completes copy 
editing; finally, SI submits to USAID on 6/6
5/16-6/6 15 3 3 - -
TOTAL 96 35 31 25 25
*Please note that this assumes a 6-day work week in the field and a 5-day work week before and after fieldwork.
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 51
X. REPORTING REQUIREMENTS 
The total pages, excluding references and annexes, should not be more than 30 pages. The 
following content (and suggested length) should be included in the report: 
Table of Contents
Acronyms
Executive Summary - concisely state the project purpose and background, key evaluation 
questions, methods, most salient findings and recommendations (2-3 pp.); 
1. Introduction – country context, including a summary of any relevant history, demography, 
socio-economic status etc. (1 pp.); 
2. The Development Problem and USAID’s Response - brief overview of the development 
problem and USAID’s strategic response, including design and implementation of the SDLG 
program and any previous USAID activities implemented in response to the problem, (2-3 
pp.); 
3. Purpose of the Evaluation - purpose, audience, and synopsis of task (1 pp.); 
4. Evaluation Methodology - describe evaluation methods, including strengths, constraints 
and gaps (1 pp.); 
5. Findings/Conclusions - describe and analyze findings for each objective area using 
graphs, figures and tables, as applicable, and also include data quality and reporting system 
that should present verification of spot checks, issues, and outcomes (12-15 pp.); 
6. Lessons Learned - provide a brief of key technical and/or administrative lessons on what 
has worked, not worked, and why for future project or relevant program designs (2-3 pp.); 
7. Recommendations – prioritized for each key question; should be separate from 
conclusions and be supported by clearly defined set of findings and conclusions. Include 
recommendations for future project implementation or relevant program designs and 
synergies with other USAID projects and other donor interventions as appropriate (3-4 pp).
Annexes – to include statement of work, documents reviewed, bibliographical documentation, 
evaluation methods, data generated from the evaluation, tools used, interview lists, meetings, 
focus group discussions, surveys, and tables. Annexes should be succinct, pertinent and 
readable. Should also include if necessary, a statement of differences regarding significant 
unresolved difference of opinion by funders, implementers, or members of the evaluation team 
on any of the findings or recommendations. 
The report format should be restricted to Microsoft products and 12-point type font should be 
used throughout the body of the report, with page margins one inch top/bottom and left/right. 
Annex-A 
List of Document (would be provided by USAID) 
1. SDLG program document 
2. SDLG PMP
3. SDLG performance report 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 52
ANNEX II: DOCUMENTS 
REVIEWED 
ARD (Associates in Rural Development) (2001) GOLD: USAID Support to Philippines Local Governance, 
Governance and Local Democracy Project 1995-2001. Compact disk should be available from 
TetraTech ARD. 
Asian Development Bank. 2012. The Urban Governance and Infrastructure Improvement 
Project in Bangladesh: Sharing Knowledge on Community-Driven Development (Manila: 
ADB, August). 
Baldwin, Kate, and Shylock Muyenga. 2014. Impact Evaluation of Supporting Traditional 
Leaders and Local Structures to Mitigate Community-Level Conflict in Zimbabwe, Final 
Report (Arlington, VA: Social Impact, August). 
Blair, Harry, Michael Calavan, Md. Azizur Rahman Siddique, and Naim Mostafa. 2012. Evaluation 
of the Improving Local Level Governance Project in Bangladesh: Combining Traditional 
Folk Arts with Democratic Local Governance (Dhaka: Social Impact, November). 
Blair, Harry. 2013. “Participatory budgeting and local governance,” in Joakim Öjendal and Anki 
Dellnas, eds., The Imperative of Good Local Governance: Challenges for the Next Decade of 
Decentralization (Tokyo: United Nations University Press), 145-178. 
Center for Urban Studies. 2011. Strengthening Democratic Local Governance: An Investigation 
into the Roles and Authorities of Local Governments in Bangladesh (Dhaka: CUS, 27 
November). 
Center for Urban Studies. 2012. Research Report on Strengthening Democratic Local 
Governance: An Investigation into the Roles and Authorities of Local Governments in 
Bangladesh (n.p., April). 
Collier, David, and James Maloney. 1996. “Insights and pitfalls: Selection bias in qualitative 
research,” World Politics 49 (1), 56-91. 
Data International. 2011. SDLG Baseline Survey (Dhaka: Data International, June). 
GOB. 2009. Local government acts for Union Parishads, Paurashavas, and Upazila Parishads, 
accessed 22 June 2014 at <<<< http://www.dwatch-bd.org/lgacts.html>>. 
Kabir, Mahfuz. 2012. Leveraging the Periphery: Effective Decentralization at Upazila and Union 
Parishad (Dhaka: US, August). 
Keshishian, Mike, and Maureen Taft-Morales. 2008. Evaluation of the USAID/Bangladesh Local 
Government Activity (n.p., 7 October). 
Khan, Mohammad Mohabbat. 2009. Decentralization in Bangladesh: Myth or Reality? (Dhaka: A H 
Development Publishing House). 
Nijera Kori. 2009. Annual Report 2008-2009 (Dhaka: Nijera Kori). Accessed at 
<http://www.nijerakori.org/documents/Annual_report_2007-2008.pdf>.
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 53
Panday, Pranab Kumar. 2011a. Working of the Upazila Parishad: Selected Case Studies (Final 
Report) (Dhaka: SDLG, 12 December). 
Panday, Pranab Kumar. 2011b. “Local government system in Bangladesh: How far is it 
decentralized?” Lex Localis – Journal of Local Self-Government 9 (3), 205-230. 
Rahman, Hossain Zillur, and S. Aminul Islam. 2002. Local Governance and Community Capacities: 
Search for New Frontiers (Dhaka: University Press Limited). 
Sabet, Daniel M., and Monzurul Haque. 2014. SDLG Endline Survey Report (Dhaka: Org-Quest 
Research, April). 
Sabet, Daniel, and Mijanur Rahman. 2013. SDLG Midline Survey Report (Dhaka: Data 
International, April). 
SDLG. 2011-2012. Work Plans, 2011, 2012, 2013 (Dhaka: SDLG). 
SDLG. 2011-2013. Strengthening Democratic Local Governance (SDLG) in Bangladesh: Semi￾Annual Progress Performance Report, Numbers 1-5 (Dhaka: SDLG). 
SDLG. 2013. Strengthening Democratic Local Governance (SDLG) in Bangladesh: Quarterly 
Progress Performance Report, Numbers 6-7 (Dhaka: SDLG). 
Siddiqui, Kamal. 2000. Local Governance in Bangladesh: Leading Issues and Major Challenges 
(Dhaka: University Press Limited). 
Sobhan, Farooq, et al. 2013. Local Government Audit and Accountability Systems: A 
Framework for Analysis (Dhaka: Bangladesh Enterprise Institute, March). 
Swiss Cooperation Office. 2009. Strategy Note for the Domain Local Governance (Dhaka: 
Swiss Cooperation Office, April). 
Swiss Development Corporation. 2013. Swiss Cooperation Strategy Bangladesh 2013-2017 
(Dhaka: SDC, November). 
Unitrend. 2012. SDLG Insight Mining: Topline Findings, version 1.1 (Dhaka: Unitrend, 
November). 
Unnayan Shamannay. 2013. Research and Policy Advocacy on Local Governance: Role of 
International Donors and Local NGOs (Dhaka: US, 10 October). 
Unnayan Shamannay. 2013. Overlapping Sources of Revenue in LGU Laws, revised (Dhaka: US, 
10 October). 
USAID. 2014. Local Systems: A Framework for Supporting Sustained Development 
(Washington: USAID, April). 
Vella, Francis. 1998. “Estimating models with sample selection bias: A survey,” Journal of 
Human Resources 33 (1), 127-169. 
World Bank. 2011. Project Appraisal Document on a Proposed Credit in the Amount of SDR 
185.80 Million (US$290 Million Equivalent) to the People’s Republic of Bangladesh for a 
Second Local Governance Support Project (31 October). 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 54
World Bank. 2013. Project Appraisal Document on a Proposed Credit in the Amount of SDR 
266.6 Million (US$410 Million Equivalent) to the People’s Republic of Bangladesh for a 
Municipal Governance and Services Project (26 November). 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 55
ANNEX III: EVALUATION 
INSTRUMENTS 
This annex consists of protocols for the team’s individual interviews (UP chairs & PS mayors; 
UZP chairs & UNOs; UP secretaries & PS executive officers) and then group interviews (UP 
and PS council members; CiG group members). 
Revised version Interviewer: AH HB NM WC 
PROTOCOL FOR INDIVIDUAL INTERVIEWS 
(UP CHAIR/VICE-CHAIR/MUNICIPAL MAYOR/DEPUTY) 
UP/PS name ______________________________ Date of interview ______March 2014 
Respondent name___________________________ 
Position_________________________ 
1. Background (how long have you been in the position, what did you do before that, etc.) 
2. Did you participate in the SDLG training workshop(s)? If so, what did you think of 
the training? Was it useful (e.g., in explaining the new local government laws, in other ways)? 
What was the most useful topic? How has SDLG done with follow-up to the training? 
Did the Program Officer make regular visits? If so, was s/he helpful with good advice and 
guidance? 
3. When did the CiG program begin here (how many budget cycles has it gone through)? 
In the time that you’ve been working with the CiG groups, what has been your experience? To 
what extent were they helpful and in what ways? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 56
4. One major task for the CiG groups has been to support the UP/PS planning process. 
How has this worked? What have they done? 
5. How have the Standing Committees been doing? Do you find them useful? Which ones 
are most important? Have the CiG groups been helpful here? 
6. Has UP/PS revenue generation improved? Has SDLG been helpful here? 
7. One SDLG objective was to improve service delivery. Do you think it has improved? If 
so, how? In what sectors (health, education, agriculture)? Can you give some examples? 
What does “transparency” mean to you? Do you think local governance has become more 
“transparent”?
8. Has the CiG process had any effect in improving candidate selection for social safety net
programs (VGD, VGF, widow’s allowance, etc.)? How about the management of these 
programs? 
9. Has women’s participation in UP/PS governance increased? If so, what are women doing 
now that they didn’t do earlier? Youth?
10. What has happened after the end of SDLG in December 2013? Does the CiG system 
seem sustainable to you? What has SDLG done to encourage it to continue? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 57
11. Is your union parishad/paurashava a member of BUPF/MAB? What is the value of these 
organizations? 
12. What do you think is the Most Significant Change that SDLG has played a role in? 
 13. Improvements. Every donor-supported program can be improved. SDLG has been a 
“pilot program” reaching about one-tenth of all UPs in Bangladesh and about one-sixth of all the 
paurashavas. If USAID (or some other donor) would like to support an expansion of the 
program to other LGUs in the future, what changes would you suggest? What improvements 
could be made in the program? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 58
Revised version Interviewer: AH HB NM WC 
PROTOCOL FOR INDIVIDUAL INTERVIEWS 
(UPAZILA PARISHAD CHAIR/VICE-CHAIR, NIRBAHI OFFICER) 
UP/PS name ______________________________ Date of interview ______March 2014 
Respondent name___________________________ 
Position_________________________ 
1. Background (how long have you been in the position, what did you do before that, etc.) 
2. Did you participate in the SDLG training workshop(s)? If so, what did you think of 
the training? Was it useful (e.g., in explaining the new local government laws, in other ways)? 
What was the most useful topic? How has SDLG done with follow-up to the training? Did 
the Program Officer make regular visits? If so, was s/he helpful with good advice and guidance? 
3. In general, would you say that the UP/PS chairs have improved their performance, now 
that they have taken the SDLG training? Do they know more about their jobs? Or is it hard 
to see much difference in them? 
4. One major for SDLG has been to support the UP/PS planning process, beginning with 
the ward shava planning meeting and continuing with the annual open budget meeting and final 
UP decisions on budget. This should have resulted in your receiving annual plans that are 
better prepared. Have you seen any improvement in these plans? 
5. Has this upazila participated in the Upazila Resource Team (URT) program, which has 
included training from NILG? If so, has it been useful? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 59
6. Has your UZP joined any of the national UZP associations? If so, what benefit do you 
see in the one you joined? What about political partisanship in these organizations? 
7. Improvements. Every donor-supported program can be improved. SDLG has been a 
“pilot program” reaching about one-fifth of all UZPs in Bangladesh. If USAID (or some other 
donor) would like to support an expansion of the program to other LGUs in the future, what 
changes would you suggest? What improvements could be made in the program? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 60
Revised version Interviewer: AH HB NM WC 
PROTOCOL FOR INDIVIDUAL INTERVIEWS 
(UP SECRETARY/PS ACCOUNTANT) 
UP/PS name ______________________________ Date of interview ______March 2014 
Respondent name___________________________ 
Position_________________________ 
1. Background (how long have you been in the position, what did you do before that, etc.) 
2. Did you participate in the SDLG training workshop(s)? If so, what did you think of 
the training? Was it useful (e.g., in explaining the new local government laws, in other ways)? 
What was the most useful topic? How has SDLG done with follow-up to the training? Did 
the Program Officer make regular visits? If so, was s/he helpful with good advice and guidance? 
3. Have you been working with the CiG groups? If so, how many budget cycles has it 
been through? What has been your experience? To what extent were they helpful and in what 
ways (e.g., tax collection)? 
4. Has UP/PS revenue generation improved? Which components have improved and by 
how much? Has SDLG been helpful here? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 61
5. What has been the balance in UP/PS revenues between grants from ADP or other 
outside sources and locally generated revenue? Has this balance changed since the SDLG 
program began? If so, how? 
6a. [if this is a union parishad]. Could you tell us about your UP’s participation in the World 
Bank’s LGSP (Local Government Support Program)? How much funding does it provide, and 
how is it used? How does their evaluation component work? Are there any bonus aspects 
for good performance? 
6b. [if this is a paurashava]. Did your paurashava participate in the UGiiP program (Urban 
Governance and Infrastructure Improvement Program) funded by the Asian Development 
Bank? If so, how much funding did the program provide, and when (phase 1 and/or phase 2)? 
How was the money used? 
7. Do citizens have access to your accounts; can any citizen see the books? Have any 
citizens asked to see the books, either members of the CiG or ordinary citizens? 
8. Improvements. Every donor-supported program can be improved. SDLG has been a 
“pilot program” reaching about one-tenth of all UPs in Bangladesh and about one-sixth of all the 
paurashavas. If USAID (or some other donor) would like to support an expansion of the 
program to other LGUs in the future, what changes would you suggest? What improvements 
could be made in the program? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 62
Revised version Interviewer: AH HB NM WC 
PROTOCOL FOR GROUP INTERVIEWS 
(UP/PS ELECTED MEMBERS) 
UP/PS name ______________________________ Date of interview ______March 2014 
Respondents names (1) _______________________ (2) 
______________________________
 
(3)__________________________________ (4) 
________________________________ 
(5) _________________________________ (6) 
________________________________ 
1. Introductions & Background (how long have you been a UP/PS member, what is your 
occupation) 
2. Did you participate in the SDLG training workshop(s)? If so, what did you think of 
the training? Was it useful (e.g., in explaining the new local government laws, in other ways)? 
What was the most useful topic? How has SDLG done with follow-up to the training? 
Did the Program Officer make regular visits? If so, was s/he helpful with good advice and 
guidance? 
3. When did the CiG program begin here (how many budget cycles has it gone through)? 
In the time that you’ve been working with the CiG groups, what has been your experience? To 
what extent were they helpful and in what ways? 
4. What do you think has been the Most Significant Change in UP/PS governance since the 
SDLG program began in 2012? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 63
5. One major task for the CiG groups has been to support the UP/PS planning process.
What have they done? 
6. How have the Standing Committees been doing? Which ones do you belong to? How 
many of them are active on a regular basis (periodic meetings, active monitoring of sectoral 
activities)? Do you find them useful? Have the CiG groups been helpful here? 
7. Participation. Would you say that ordinary citizens have become more engaged in local 
government since SDLG began its work? If so, how? Participation in ward shava meetings? 
In the open budget meeting? What kind of participation (entering the discussions, or just 
attending)? 
8. Are ordinary citizens making more demands on you and on the UP than before? If so, 
of what kind? 
9. Has UP/PS revenue generation improved? Has SDLG been helpful here? 
10. One SDLG objective was to improve service delivery. Do you think it has improved? If 
so, how? In what sectors (health, education, agriculture)? What does “transparency” mean to 
you? Do you think local governance has become more “transparent”?
11. Has the CiG process had any effect in improving candidate selection for social safety net
programs (VGD, VGF, widow’s allowance, etc.)? How about the management of these 
programs? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 64
12. Has women’s participation in UP/PS governance increased? If so, what are women 
doing now that they didn’t do earlier? Youth?
10. What has happened after the end of SDLG in December 2013? Does the CiG system 
seem sustainable to you? What has SDLG done to encourage it to continue? 
14. Improvements. Every donor-supported program can be improved. SDLG has been a 
“pilot program” reaching about one-tenth of all UPs in Bangladesh and about one-sixth of all the 
paurashavas. If USAID (or some other donor) would like to support an expansion of the 
program to other LGUs in the future, what changes would you suggest? What improvements 
could be made in the program? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 65
Revised version Interviewer: AH HB NM WC 
PROTOCOL FOR GROUP INTERVIEWS 
(CITIZENS-in-GOVERNANCE MEMBERS) 
UP/PS name ______________________________ Date of interview ______March 2014 
Respondents names (1) _______________________ (2) 
______________________________ 
 
(3)__________________________________ (4) 
________________________________ 
(5) _________________________________ (6) 
________________________________ 
1. Introductions & Background (how long have you been a CiG member, what is your 
occupation?) 
2. Did you participate in the SDLG training workshop(s)? If so, what did you think of 
the training? Was it useful (e.g., in helping you to become a good CiG member, in explaining 
what LGUs are supposed to do)? What was the most useful topic? How has SDLG done 
with follow-up to the training? Did the Program Officer make regular visits? If so, was s/he 
helpful with good advice and guidance? 
3. When did the CiG program begin here (how many budget cycles has it gone through)? 
In the time that you’ve been working with the CiG groups, what has been your experience? To 
what extent were they helpful and in what ways? 
4. What was your experience in working with the UP/PS? Were the chair and the members 
welcoming right away, or did it take some time to get integrated with them? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 66
5. Do you think the CiG participants have been able to make an impact on how the UP/PS 
does it work? If so, how would you describe that impact? 
6. What do you think has been the Most Significant Change in UP/PS governance since the 
SDLG program began in 2012? 
7. One major task for the CiG groups has been to support the UP/PS planning process, 
beginning with the ward shava planning meeting and continuing with the annual open budget 
meeting and final UP decisions on budget. How has this worked? 
7. How have the Standing Committees been doing? Are you a member of one or more of 
them? How many of them are active on a regular basis (periodic meetings, active monitoring
of sectoral activities)? Do you find them useful? Do you think the CiG members have made 
an impact with them? How? 
8. Participation. Would you say that ordinary citizens have become more engaged in local 
government since SDLG began its work? If so, how? Participation in ward shava meetings? 
In the open budget meeting? What kind of participation (entering the discussions, or just 
attending)? 
9. Has UP/PS revenue generation improved? Has SDLG been helpful here? 
10. One SDLG objective was to improve service delivery. Do you think it has improved? If 
so, how? In what sectors (health, education, agriculture)? Have public servants become more 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 67
“accountable” to the citizenry? What does this term “accountable” mean to you? What does 
“transparency” mean to you? Do you think local governance has become more “transparent”?
11. Has women’s participation in UP/PS governance increased? If so, what are women 
doing now that they didn’t do earlier? Youth?
12. Has the CiG process had any effect in improving candidate selection for social safety net
programs (VGD, VGF, widow’s allowance, etc.)? How about the management of these 
programs? 
13. What has happened after the end of SDLG in December 2013? Does the CiG system 
seem sustainable to you? What has SDLG done to encourage it to continue? 
 
14. Improvements. Every donor-supported program can be improved. SDLG has been a 
“pilot program” reaching about one-tenth of all UPs in Bangladesh and about one-sixth of all the 
paurashavas. If USAID (or some other donor) would like to support an expansion of the 
program to other LGUs in the future, what changes would you suggest? What improvements 
could be made in the program? 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 68
ANNEX IV: INTERVIEW LIST 
Dhaka interviews 
Organization Individuals 
interviewed
Titles
USAID Mission Rumana Amin
Sherina Tabbasum
Jason Smith
Billy Woodward
COR, BDGPE
COR, SDLG
DG Deputy Director
DG Officer
SDLG Jerome Sayre
Zarina Rahman Khan
Abu Md Mohsin
COP, SDLG
Deputy COP, SDLG
Team leader, Citizen Participation
MAB Shahmim Al Razi
Md. Azmat Ullah Khan
Secretary General, MAB
President, MAB
ADB M. Rafiqul Islam
Elma Morsheda
Senior Project Officer
Project Officer
JICA-UDCC Arika Munekata Advisor, Local Governance
DFID Richard Butterworth Senior Governance Advisor
GiZ Hans Joachim Hermann Principal Advisor
World Bank Zahed H. Khan
Christopher T. Pablo
Senior Urban Specialist
Senior Urban Development Officer
DFATD (Canadian 
High Commission)
Peggy Thorpe First Secretary (Development)
TAF Hasan Mazumdar
Russell Pepe
Country Representative
COP, PRODIP
Unnayan Shamannay Shaheen ul Alam
Mahfuz Kabir
Project Coordinator
Consultant
SDC/Sharique Sohel Ibn Ali
Lilia Tverdun
Sharique Project Director
SDC Program Officer
Centre for Urban 
Studies
Prof. Nazrul Islam Chairman
Democracy Watch Taleya Rehman
Wazed Feroj
Executive Director
Deputy Executive Director
National Institute for 
Local Government
Kabir M. Ashraf Alam Director General
Data International Najmul Hossain
A F. M. Azizur Rahman
Managing Director
Director
UzPAB Faizur Rahman Fakir President
BUPF Mahbubur Rahman Tulu President
UNDP Shaila Khan
Md Mozammel Haque
Tofail Ahmed
Md. Sydur Rahman Molla
Assistant Country Director, Local Governance
Project Manager, Upazila Governance Project
National Technical Advisor
Programme Analyst
Ministry LGRD&C Mozammel Haq
Ashok Madhab Rao
Additional Secretary, LGSP II
Additional Secretary, Municipalities
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 69
Organization Individuals 
interviewed
Titles
Shamima Nargis
Mohammad Yah-ya￾Bhuian
Ajit Kumar Paul
Joint Secretary, Local Gov Div
Deputy Secretary & Deputy Project Director, 
LGSP II
Sr. Local Governance Audit Specialist, LGSP II
Ministry of Liberation 
War
Hon A. K. M. Mozammel Minister (former Executive Committee member, 
MAB)
SDC Melina Papageorgios 
Trippolini
Lilia Tverdun
Sohel Ibne Ali
Program Manager, Local Governance Portfolio
Director, Local Governance Program Sharique
Program Manager, Local Governance
RTI/SGLG Azmal Hossain Former Program Officer, SGLG
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 70
Field interviews 
Organization Individuals 
interviewed Titles
Kolaroa PS Saiful Islam PS Secretary
Kolaroa PS Aktarul Islam Mayor
Kolaroa UZP Aminul Islam Vice Chairman
Kolaroa PS Rafiqul Islam
Razu
Jamil
PS Council Members
Shushila PNGO, 
Satkhira PS
Mostafa Aktheruzzaman
Mahabubul Alam Minto
Sumon Chatterjee
S.Abul Haranat
Sushan Monelal
Ahmed Sharif
Moniruzzaman
Golam Mostofa
Deputy Director & Program Coordinator
Project Officer
Project Officer
M&E Officer
Program Assistant
Program Officer
Sr. Program Officer
Program Manager
Kerakata UP Rafiqul Islam
Ismail Hossain
Nurul Amin
Mizanur Rahman
CiG Male Members
Keralkata UP Najma
Momtaz
CiG Female Members
Keralkata UP AsgarAli
Rabiul Islam
Shohidul Islam
Ashraf Ali
UP Male Council Members
Keralkata UP Nur Jahan Begum
Sonia Lila
UP Female Council Members
Keralkata UP Morshed Ali UP Chairman
Kolaroa Upazila Anup Kumar Talukdar Upazila Nirbahi Officer
Kolaroa Upazila Nazrul Islam Upazila Chairman
Kolaroa Upazila Aminul Islam Laltu Upazila Vice Chairman
Kolaroa PS Aktarul Islam Mayor
Kolaroa PS Saiful Islam Secretary
Kolaroa PS Ishaque Ali
Abdul Khaleque
Sheikh Faruk Ahmed
Ranajit Ghosh
CiG Male Members
Kolaroa PS Rashma Khatun
Kalima Khatun
CiG Female Members
Kolaroa PS Zamil Hossain
Rafiqul Islam
Raju
Male Council Members
Jibannagar Golam Mortuza
Sarhana Akter Rini
UZP Chair
UZP Vice Chair
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 71
Organization Individuals 
interviewed Titles
Chuadanda PS Ramjan Ali
Jahangir Alam Malik
Shahidul Kador Joadar
Md Sirajul Islam Monir
Md Nazmux Salahin Liton
Md Faraz Ali Sheikh
PS Male Council Members
Chuadanga PS Shakil Ahmen
Md Abdul Kuddus
Masud Hossain Jourdar
CiG Male Members
Chuadanga PS Sultana Ara
Jahamra Begum
Selina Yasmin (Shompa)
CiG Female Members
Chuadanga PS Saiful Arif Biswaw Acting Mayor
Chuadanga PS Kazi Shoruful Islam PS Secretary
Uthali UP Younus Ali Panel Chairman
Uthali UP Liakat Ali Secretary
Uthali UP Aminul Islam
Abdul Kuddus
Bazlur Rahman Tutu
Mahbub Uddin
Dalu Uddin
Abdul Hannan
UP Council Male Members
Uthali UP Morium UP Council Female Member
Uthali UP Beauty
Shahina Akter
CiG Female Members
Uthali UP Abu Jafor
Nurul Islam
Bazlur Rahman
Hafizur Rahman
CiG Male Members
Wave Foundation, 
PNGO SDLG
Abdus Salam
Akhtaruzzaman
Assistant Project Coordinator
Project Officer
Democracy Watch 
PNGO, Jessor
Dulal Hossain
Md Mujibur Rahman
Parimal Karmokar
Layla Akter
M&E Officer
Program Officer
Program Officer
Program Officer
Sahjadpur PS Golam Azam (Mawla) UZP Vice Chair & Acting Chair
Sahjadpur Upazila Rasel Sabrim Upazila Nirbahi Officer
Sahjadpur PS Najbul Islam Mayor
Sahjadpur PS Shafiqul Islam - male
Zabir Hussain – male
Zakra Parvin - female
CiG Members
Sahjadpur PS Liaquat Hossain
Pradip Kumar Poddar
Md Yunus Ali
PS Council Members
Sahjadpur PS Beljur Rahman Khan PS Secretary
Kayampur UP Zainul Alam Chairman
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 72
Organization Individuals 
interviewed Titles
Kayampur UP Aklima Khatun UP Female Council Member
Kayamput UP Abul Hasan
Zahidul Islam
Zakir Hossain
Habibur Rahman
UP Male Council Members
Kayampur UP Sajeda Pervin CiG Female Member
Kayampur UP Aktar Hossain
Ramjan Ali
CiG Male Member
Manab Mukti 
Sangostha (MMS), 
PNGO
Md Habibulla Bahar
Gupal Chandra Shil\
Shamima Akhter
Nusrat Jahan
Md Fariduddin
Md Abdul Halim
Shahanar Begum
Shuma Khatum
Director
Program Coordinator
M&E Officer
Program Assistant
Finance & Administration
Program Officer
Program Officer
Program Officer
Garadaha UP Md Saiful Islam UP Chair
Garadaha UP Humayun Kabir UP Secretary
Garadaha UP Abdullah
Md Azahar Ali
Md Ayub Ali
Mapul Rana
UP Council Members
Garadaha UP Ilias Hossain CiG Male Member
Garadaha UP Champa Rani Dev
Chama Rani Dev
CiG Female Members
Garaadaha UP Monerwara Khatun
Zahera
UP Female Council Members
Pabna PS Jahurul Islam PS Council Member
Pabna PS Kampul Hasan Mayor
Pabna PS Ahmed Kabir
Sajjad Hossain
Nazmul Islam
Alamgir Hossain
CiG Male Members
Pabna PS Farhad Joarden
Md Razibul Hasan
Md Shahidul Islam
Ahraf Pramanik
Ayub Ali Sardar
PS Council Male Members
Pabna PS Afroja Khatun Chhabi
Toma Islam (Pushpa)
Laily Begum
Sardina Akter Shahana
PS Council Female Members
Pabna PS Md Sayedul Islam PS Secretary
RDRS, PNGO Md Atiqur Rahman Program Officer
Pabna PS Ray Hanna Islam Upazila Nirbahi Officer
Barara UP, Pabna Abu Sayed Khan UP Chair
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 73
Organization Individuals 
interviewed Titles
Barara UP Rejaul Karim
Shofiqur Rahman
Belal Hossain
CiG Male Members
Barara UP Champa
Sajida
Sanjida Akhter
CiG Female Members
Barara UP Rashida Khatun
Jayda Khatun
Ashiya Khatun
Female Council Members
Barara UP Shamsul Sardar
Halimuzzaman
Abul Kalam
Abdus Sobhan
Kamruzzaman Kuddus
Male Council Members
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 74
ANNEX V: TABLES AND FIGURES 
Figure 1: SDLG Goals and Components 
Table 1 Field Visit Plan 
Date Division District Upazila Paurashava Union
2-3 Mar Khulna Satkhira Kolaroa* Kolaroa* Keralkata*
4 Mar Khulna Chuadanga Jibannagar* Jibannagar
Chuadanga*
Utholi*
9 Mar Rajshahi Sirajganj Sahjadpur* Sahjadpur* Kayampur*
10 Mar Rajshahi Sahjadpur Sahjadpur Sahjadpur Garadaha Rupati*
11 Mar Rajshahi Pabna Pabna Sadar* Pabna* Barara*
*LGUs visited by the evaluation team 
Also interviewed four PNGOs 
Component 3 –
Transparent and Effective 
Service Delivery by Local 
Governments
Component 2 – Advocacy 
and Capacity Building of 
Local Government 
Associations 
Component 1- Roles 
and Authorities of Local 
Governments
Strategic Objective: More effective and responsive democratic institutions and practices
Component 4 –
Citizen Participation 
in Local Decision 
Making
Component 5- Windows of Opportunity 
(Crosscutting)
Program Goal: Improve transparency and participatory public administration at the sub-national level and enhance legal 
and policy reform at the national level in order to promote and expand decentralization 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 75
Figure 2: SDLG Sample LGUs Visited 
Chuadanga
(1 UZP, 1 PS, 1 UP)
Satkhira
(1 UZP, 
1PS, 1 UP)
SiSirajganj
(1 UZP, 1 PS, 3 UPs)
Pabna
(1 UZP, 1 PS, 1 UP)
Table 2: SDLG Evaluation Questions and Conclusions in Brief 
Question Sub-question Components 1 & 2 Components 3 & 4 
1. Effectiveness 1a. Results Repetitive research
Little institution building 
Mostly good PMP
T/C results mixed 
1b. Surprises Strong LGA pushback Enthusiasm level 
1c. Added value US macro-level lobbying 
model.
RevGen + CiG model 
2. Efficiency Alternatives Long-term USAID strategy Longer LOP
Confrontational advocacy 
3. Management 
& 
administration–
working with: 
3a. SDLG partners Conflicts with LGAs PNGOs good but one-off 
3b. Other donors Little linkage Little linkage 
3d. LGU elected Little LGA work with its 
members 
Well done 
3e. Citizens Little involvement CiG good 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 76
Question Sub-question Components 1 & 2 Components 3 & 4 
3f. GOB macro￾level 
No sustained engagement Not part of project design 
4. Sustainability 4a. Program 
activities 
MAB may survive
BUPF & UZPAB less likely 
CIG enthusiasm but little 
institutionalization 
4b. Possible 
measures 
Links to other LGAs Longer LOP 
5. Relevance 5a. To present dev 
situation 
Little technical depth CiG an excellent model 
5b. To future 
programming 
Needs revamped strategic 
vision 
CiG model should be spread 
6. Cross-cutting 6a. Gender Some mention Legal & SDLG reqts met
Small leadership programs 
6b. Youth Little attention given Little attention given 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 77
Table 3: SDLG Spending on Support to LGAs March 2011 to December 2013 (in 
US$) 
LGA Core Costs 
Secretariat
Program Costs -
Direct
Program Costs –
Grants
Total
MAB 58,130 149,383 39,402 246,915 
BUPF 44,355 281,798 16,275 342,428 
UzPAB 20,556 82,180 - 102,736 
Subtotal 123,041 513,361 55,677 692,079 
Source: SDLG 
Table 4: Comparison of the Percent of Union Parishad Respondents who Report 
Being Aware of UP Efforts to Raise Revenue (PMP 4.2) 
Baseline Endline Treatment 
difference
Control 
difference
Group 
Treatment Control Treatment Control difference
% % % % % % %
Aware of 
efforts
4.1 4.5 11.6 4.1 7.5 -0.4 7.9
Unaware of 
efforts
31.0 49.6 29.6 33.6
Don’t know 64.9 45.9 58.8 62.3
Base 
(Unweighted)
921 359 750 300
Source: Sabet and Haque (2014). 
Table 5: Comparison of the Percent of Paurashava Respondents who Report Being 
Aware of PS Efforts to Raise Revenue (PMP 4.2) 
Baseline Endline Treatment 
difference
Control 
difference
Group 
Treatment Control Treatment Control difference
% % % % % % %
Aware of 
efforts 4.8 15.1 10.3 16.0 5.5 0.9 4.6
Unaware of 
efforts 31.1 31.1 28.0 23.8
Don’t know 64.2 53.8 61.8 60.2
Base 
(Unweighted) 479 81 288 120
Source: Sabet and Haque (2014) 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 78
Table 6: Comparison of the Percent of Union Parishad Respondents who Agree 
that the UP Manages their Funds with Transparency and Accountability (PMP 4.2) 
Baseline Endline Treatment 
difference
Control 
difference
Group 
Treatment Control Treatment Control difference
% % % % % % %
Yes 25.1 21.9 48.2 37.3 23.1 15.4 7.7
No 25.0 29.0 28.6 35.9
Don't know 50.0 49.1 23.2 26.8
Base 
(Unweighted) 898 349 553 228
Source: Sabet and Haque (2014). 
Table 7: Comparison of the Percent of Paurashava Respondents who Agree that 
the PS Manages its Funds with Transparency and Accountability (PMP 4.2) 
Baseline Endline Treatment 
difference
Control 
difference
Group 
Treatment Control Treatment Control difference
% % % % % % %
Yes 17.6 16.3 57.3 57.9 39.7 41.6 -1.9
No 23.0 18.6 21.7 19.1
Don't know 59.3 65.1 21.1 23.0
Base 
(Unweighted) 480 81 193 87
Source: Sabet and Haque (2014). 
Table 8: Own Source Revenue Expenditure Patterns in 2012-13
LGU Type Salaries Development 
Objectives
Office et 
al. Unspent Total
PSs 27% 31% 28% 14% 100%
UPs 24% 50% 29% 6% 100%
Source: SDLG. 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 79
Table 9: Comparison of Citizens’ Evaluations of the Quality of Road Construction, 
Repair, and Maintenance in Union Parishads (PMP 3.5) 
Baseline Endline Treatment 
difference
Control 
difference
Group 
difference Treatment Control Treatment Control
% % % % % % %
Poor/Very poor 11.3 13.5 6.5 2.9
Moderate 35.9 38.9 36.7 43.3
Good/Excellent 52.8 47.7 56.9 53.1 4.1 5.4 -1.3
Base 
(Unweighted)
879 342 603 171
Total sample size 921 359 750 300
Source: Sabet and Haque (2014). 
Table 10: Comparison of Citizens’ Evaluations of the Quality of Maintaining 
Market Places for Citizens in Paurashavas (PMP 4.10) 
Baseline Endline Treatment 
difference
Control 
difference
Group 
difference Treatment Control Treatment Control
% % % % % % %
Poor/Very poor 6.1 5.7 5.5 2.9
Moderate 45.3 36.8 39.6 17.0
Good/Excellent 48.5 57.6 54.9 80.0 6.4 22.5 -16.1
Base 
(Unweighted)
352 57 80 44
Total sample 
size
480 81 288 120
Source: Sabet and Haque (2014). 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 80
ANNEX VI: PERFORMANCE MANAGEMENT 
PLAN FOR SDLG: TARGETS AND ACTUAL 
RESULTS, 2011-2013 
SDLG Performance Indicators – Results of 2011 (Year 1), 2012 (Year 2) and 2013 (Year 3) 
Indicator 
Definition
Treatment/
Control Baseline Target 
Yr. 1
Actual 
Yr. 1
Target Yr. 
2
Actual
Yr. 2 Target Yr. 3 Actual
Yr. 3
LOP 
Target
LOP 
Actual
Component 1- Roles and Authorities of Local Governments
1.1 } Number of 
LGUs where 
information on local 
government services 
is readily available to 
community 
members (Citizen 
Charter)
Treatment 0 0 0 500 394 500* 499 500
1.2 } Percentage of 
elected council 
officials and civil 
servants with 
increased 
understanding of 
local government 
issues and 
capabilities
Treatment/
Control
4% of 
Sample
4%
Baseline 4% 20% of 
Sample 22.48% 40% of Sample 51% 40% of 
Sample
1.3 } Number of 
policy dialogue 
roundtables or 
research 
conferences on local 
government held 
regionally or 
Treatment 0 1 1 7 12 4 4 12
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 81
Indicator 
Definition
Treatment/
Control Baseline Target 
Yr. 1
Actual 
Yr. 1
Target Yr. 
2
Actual
Yr. 2 Target Yr. 3 Actual
Yr. 3
LOP 
Target
LOP 
Actual
nationally
1.4} Number of 
advocacy actions 
conducted by 
women elected 
representatives in 
local government
Treatment 0 4 4 6 7 8 1 18
Component 2 – Advocacy and Capacity Building of Local Government Associations 
2.1} Percentage 
increase of BUPF, 
MAB and Upazila 
Association’s own 
revenue**
Treatment
MAB
Tk500,000 
includes 
contributio
ns
0%
Baseline
0%
Baseline
100%
Tk1,000,000
+ 270%
Tk1,842,000 
includes 
contributions
(Tk404,000 in 
Dues)
150%
Tk1,250,000
+ 70.4%
Tk852,000 
in Dues
MAB=150%
Tk1,250,000
Treatment
BUPF
Tk50,000 
includes 
contributio
ns
0%
Baseline
0%
Baseline
BUPF=3,000
%
Tk1,550,000
+ 1077%
Tk588,500 in Dues
BUPF=8000% 
Tk4,050,000
+ 1968%
Tk1,034,000 in Dues
BUPF=8,000
% 
Tk4,050,000
Treatment
Upz Assoc. Upz=Tk0 0%
Baseline
0%
Baseline
50% Dues
Tk723,000 Tk155,550 in Dues 70% Dues
Tk1,012,200
21.4% Dues
Tk309,000 in Dues
70% Dues
Tk1,012,200
2.2 } Total number 
of BUPF and MAB 
members** 
Treatment
MAB= 200
BUPF= 
3,000
Baseline Baseline
MAB= 275
BUPF= 
3,500
MAB-308 
(67 dues paid)
BUPF=593
MAB= 300
BUPF= 4200
MAB = 308
-136 dues paid
BUPF = 1034
MAB 300
BUPF= 4,200
2.3 } Number of 
local and non￾governmental and 
public sector 
associations 
supported with USG 
assistance 
Treatment 0 2 2
3
(2* + 1) 3 3* 3 3
2.4 } Number of 
legal challenges to 
GOB limits on local 
gov’t authority
Treatment 1 1 0 2 4 1 1 4
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 82
Indicator 
Definition
Treatment/
Control Baseline Target 
Yr. 1
Actual 
Yr. 1
Target Yr. 
2
Actual
Yr. 2 Target Yr. 3 Actual
Yr. 3
LOP 
Target
LOP 
Actual
Component 3 – Transparent and Effective Service Delivery by Local Governments
3.1} Number of sub￾national govt 
entities receiving 
USG assistance to 
improve their 
performance
Treatment 0 0 0 600
526
(including 26 
Upazila)
600*
592
(including 92 
Upazila)
600
3.2} Number of sub￾national govts 
receiving USG 
assistance to in￾crease their annual 
own-source 
revenues (OSR)
Treatment 0 0 0 500 500 500* 500 500
3.3 } Number of 
individuals who 
received USG￾assisted training, 
including mgt skills 
and fiscal mgi, to 
strengthen local 
govt and/or 
decentralization 
Treatment 0 4,008 4,426
(revised) 24,610 23,207
25,900
(23,500*
+ 1,500)
35,058 30,118
3.4} % increase in 
tax revenue 
generated in 
targeted LGUs (own 
source revenue –
OSR)
Treatment /
Control
Avg. OSR
Union –
220,000 / 
220,000
Muni –
18,500,000 
/ 
18,650,000
Baseline Baseline 30% above 
baseline
Avg. OSR
Union￾300,274/159,173
Mean: 21.7% incr.
Median: 20.5% 
Incr.
Muni￾11,072,225/1,049,2
78
Mean: 43.8% incr.
Median: 58.3% 
incr.
50% above 
baseline
Avg. OSR
Union￾698,164/488,453
Treatment: +217.3%
Control: +122.0%
Muni￾40,078,165/15,017,6
21
Treatment: +116.6%
Control: -19.4%.
50% above 
baseline
3.5} Percentage 
increase in citizen 
satisfaction with 
selected LGU 
Treatment
% Good or 
Excellent
Union-47% 
(local 
Baseline Baseline 15% above 
baseline
% Good or 
Excellent
Union-69.5%
(local roads)
30% above 
baseline
% Good or 
Excellent
Union-56.7%
(local roads)
30% above 
baseline
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 83
Indicator 
Definition
Treatment/
Control Baseline Target 
Yr. 1
Actual 
Yr. 1
Target Yr. 
2
Actual
Yr. 2 Target Yr. 3 Actual
Yr. 3
LOP 
Target
LOP 
Actual
services roads)
Muni-37%
(local 
markets)
Result: 17.0% incr.
Muni-67.8%
(local markets)
Result: 19.5% incr.
Result: +10.0%
Muni-55.1%
(local markets)
Result: +18.1%
Component 4 – Citizen Participation in Local-Decision Making
4.1} Number of 
Citizens’ Forums 
established 
Treatment 0 0 0 500 499 500* 500 500
4.2} Percentage of 
citizens that are 
better informed 
about LGU revenue 
generation capacity 
and fiscal 
transparency.
Treatment /
Control
% Yes
Revenue 
Generation
Union￾5%/4%
Muni￾5%/18%
Fiscal 
Transp.
Union￾25%/19%
Muni￾17%/25%
Baseline Baseline 15% above 
baseline
% Yes
Revenue 
Generation
Union-19.9%/9.7%
Muni-25.3%/32.7%
Increase in 
Union:15%/4.70%
Muni:19.7%/15.4%
Fiscal Transp.
Union-41%/48%
Muni-47%/36.7%
Increase in 
Union:15.1%/25.2
%
Muni:28.7%/19.5%
30% above 
baseline
% Yes
Revenue 
Generation
Union-10.3%/4.0%
Muni-10.0%/15.0%
Increase in 
Union:5.3%/0.0%
Muni:5%/ -3%
Fiscal Transp.
Union-48%/37%
Muni-58%/55.2%
Increase in 
Union:23.0%/18.0%
Muni:41.0%/30.0%
30% above 
baseline
4.3} Participatory 
strategic planning 
and budgeting 
implemented in 
targeted LGUs
Treatment 0 0 0 500 247 600
(500* +100)
498
(including 8 Upazila) 
600
4.4} Number of 
local mechanisms 
supported with USG 
assistance for 
citizens to engage 
their sub-national 
government.
Treatment 0 0 0 3 5
5
(3* + 2) 5 5
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 84
Indicator 
Definition
Treatment/
Control Baseline Target 
Yr. 1
Actual 
Yr. 1
Target Yr. 
2
Actual
Yr. 2 Target Yr. 3 Actual
Yr. 3
LOP 
Target
LOP 
Actual
4.5} Percentage of 
citizens that feel 
their input and 
feedback was 
considered in local 
government 
decision making 
process
Treatment /
Control
% 
Somewhat 
or 
Very Much
Union- 34% 
/ 30%
Muni- 29% 
/ 26%
Baseline Baseline 15% above 
baseline
% Somewhat or
Very Much
Union-99.4%
Muni- 99.4%
*For increase, see 
the footnote on 4.5 
below
30% above 
baseline
% Somewhat or 
Very Much
Union- 90% / 86%
Muni- 91% / 89%
Increase in 
Union:56.0%/56.0%
Muni:62.0%/63.0%
30% above 
baseline
**For MAB, annual dues are collected on a July-June fiscal year basis. Membership includes all municipalities but only some pay MAB dues. 
** For BUPF and UzPAB, annual dues are collected on a calendar year basis. Membership is counted only for dues paying members. 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 85
ANNEX VII: FOOTNOTES ON 
SURVEY BASED INDICATOR 
VALUES FOR 2012 (3.4, 3.5, 4.2, AND 
4.5)
The PMP indicators accumulate data from two sources: 1) implementing partners and 2) 
surveys (baseline, midline, and LGCI). There are four specific indicators (3.4, 3.5, 4.2, and 4.5) 
of which targets and actuals come from the baseline and midline surveys. To determine Year 1 
values for the abovementioned indicators, the evaluation team used data from the baseline 
survey, and for Year 2 values, the team used midline survey data. However, after analysis of the 
midline findings, it impossible to keep the same values because the list of comparative LGUs for 
the baseline and midline changed due to technical error (see tables below). Thus, the value of 
paurashavas and union parishads in the baseline increased by three and ten, respectively due to 
this error. 
LGU type Baseline Midline Reason for changes
Paurashava (PS) 20 17 Three PSs were removed as they were not recipients of 
the treatment as intended during the baseline. These 
three PSs are in fact part of the control group.
Union Parishad 
(UP)
80 70 Ten Ups were removed as they were not recipients of 
the treatment as intended during the baseline. These ten 
UPs are in fact part of the control group.
Total 100 87
The changing baseline values: 
Indicator 3.4 
Type of LGUs Changing values Midline values Result
Treatment UP 246,748 300,274 21.7% increase
Control UP 193,582 159,173 -17.8% increase
Treatment Muni. 7,699,576 11,072,225 43.8% increase
Control Muni. 981,729 1,049,278 6.9% increase
Indicator 3.5 
Type of LGUs Changing values Midline values Result
Treatment Union Good and Excellence
52.5% local roads
69.5% local roads 17.0% increase
Treatment Muni. 48.3% local markets 67.8% local markets 19.5% increase
Indicator 4.2 
Indicator Changing values
(Treatment/Control)
Midline values
(Treatment/Control)
Result
Revenue 
Generation
Union-4.9%/5.0%
Muni-5.6%/17.3%
Union-19.9%/9.7%
Muni-25.3%/32.7%
Increase￾Union-15.05%/4.7%
Muni-19.7%/15.4%
Fiscal Transparency Union-26.5%/22.9%
Muni-18.3%/17.3%
Union-41.6%/48.1%
Muni-47.0%/36.7%
Increase￾Union- 15.1%/25.2%
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 86
Muni-28.7%/19.5%
Indicator 4.5 
Type of LGUs Changing values Midline values Result
Treatment % of somewhat and very 
much
Union-90.8%
Muni- 91.6%
Union- 99.4%
Muni- 99.4%
Increase￾Union-8.6%
Muni-7.8%
Control Union-89.2%
Muni -75.0%
Union-100%
Muni-100%
Increase￾Union-10.8%
Muni-25%
Special Note for Indicator 4.5 
Indicator 4.5 was originally based on the question, “If there is no Open Budget or Participatory 
Planning Meeting, how well do you think your views and input are considered by the 
UP/paurashava during its decision making?” The target for this indicator was set at 15 percent 
increase in the percent of citizens who responded “very much” or “somewhat” rather than 
“not at all” among midline respondents. Unfortunately, a change in the instrument made
comparison between the baseline and the midline impossible. Perhaps more importantly, 
however, this question would have offered very limited means to measure project impact. 
Individuals who had developed an appreciation for the SDLG-promoted open budget and 
planning meetings as a result of the intervention might have actually been less likely to state 
“very much” or “somewhat” rather than more likely.28
 
As a result of these two methodological problems, the evaluation team selected an alternative 
variable that offers a better measure of the intended concept of “voice.” Respondents who 
were aware of planning/ward were asked, “Do you think that citizens can voice their views and 
opinions in ward/planning meetings?” This was then repeated for those aware of open budget 
meetings. While these questions offer a better indicator, they still present a methodological 
challenge, as only a reduced percentage of respondents were aware of such meetings and 
therefore able to express an opinion of them. As such, to increase the sample size, the team 
combined respondent evaluations of both the planning/ward and open budget. This allows the 
team to be fairly confident in the resulting percentages among union parishad respondents and 
treatment paurashava respondents; however, the evaluation team still cannot be confident in 
the percentages in the paurashava control areas. As a result of the reduced sample sizes, in the 
tables that follow, we present both the percentages and the frequencies. 
 
28 There is good reason to think that this would be the case. In the baseline, only 34 percent of respondents in the 
treatment group (and 30 percent of respondents in the control group) felt that their input was very much or 
somewhat considered in the absence of planning/ward and open budget meetings, a very poor evaluation 
compared to the generally positive evaluations provided by survey respondents of their public officials. In short, it 
seems likely that the question was inadvertently asking people to evaluate the new public participation mechanisms 
rather than the openness of public officials. 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 87
The vast majority of union parishad respondents aware of either planning/ward meetings and 
open budget meetings felt that they could somewhat voice their views and opinions in these 
venues. Together, 99.4 percent of midline treatment respondents felt somewhat or very much 
that they were able to voice their views, compared with 90.8 percent of baseline respondents, a 
difference of 8.6 percentage points. Nonetheless, a similar change was observed in the control 
group, and as such, the team is not able to conclude with sufficient confidence that the 
intervention had an impact on perception of the ability to voice views. 
. 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 88
ANNEX VIII. EVALUATION DESIGN MATRIX – 26 
FEBRUARY 2014 
DATA COLLECTION STRATEGIES--In Response to Evaluation Questions 
Evaluation
Question
Assumptions/Com
ments Field Visit Dhaka Interviews Document Review
1. Effective￾ness: To what 
extent has the SDLG 
program been 
successful in
achieving its planned
objectives? 
This is the core 
query of the 
evaluation.
Our field visits will provide the main 
data to answer this ques-tion. Because 
we will visit UPs, PSs and UZPs, the 
sample set will be small and stretched, 
and the data gathered will neces-sarily 
be qualitative, but there will be enough 
detail from a broad spectrum of elected 
officials, govt officers, LG association 
personnel, and citizens to provide a 
good account.
Dhaka interviews will give us a 
“big picture” here. USAID, 
SDLG, GOB, BUPF/MAB 
personnel will all be interviewed 
at some length.
The survey data will 
be useful in gauging 
progess on SDLG 
components 3 and 
4.
SDLG’s periodic 
reports & other 
documents will 
provide details. 
Were there any 
unexpected 
outcomes? 
Their absence 
would be surprising.
Local respondents will provide 
examples.
Other donor experience should 
give insight on this.
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 89
What added value 
has USAID brought 
to the local 
governance sector in 
Bangladesh through 
SDLG?
The question asks 
what value has 
SDLG added to the 
art of strengthening 
local govern-ance, 
not just to 
governance in the 
LGUs supported.
Local respondents will have plenty to 
say about this. The question also 
relates to sustainability (#4): will any 
added value endure beyond EOP?
Institutional memories at this 
level (esp. USAID and GOB
personnel) will probably be 
more reliable than those at local 
level,
SDLG reports will 
show project 
outputs, enabling us 
to look for 
outcomes.
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 90
Evaluation
Question
Assumptions/Comm
ents Field Visit Dhaka Interviews Document Review
2. Efficiency:
Is there evidence 
from the 
implementation of 
SDLG to suggest 
that alternative 
program approaches 
may have been more 
successful or was 
SDLG’s approach 
the most efficient?
Bangladeshis love 
to speculate, so it 
will not be hard to 
elicit ideas here.
Other donor LG programs 
encountered on our own field 
visits would provide ideas, but this 
would be serendipitous. 
Asking our own respondents “what 
could SDLG have done better” will 
elicit many responses.
Other donors presently 
involved in LG programs 
(esp. World Bank, UNDP, 
SDC, JICA) have each used 
their own approaches, which 
we will compare with SDLG. 
If time permits, other donors 
with past experience could 
also be queried (e.g., DfID, 
DANIDA).
GOB officials will also have 
ideas on this topic. 
We expect little 
attention to “paths 
not taken” in the 
documents we will 
review.
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 91
Evaluation
Question
Assumptions/Comment
s
Field Visit Dhaka Interviews Document Review
3 Manage￾ment & Ad￾ministration:
How effective 
and flexible has 
the SDLG 
management 
been in working 
with (a) 
implement-ing 
partners, 
Answers would come 
both from SDLG staff 
and the partners.
Partners may show 
some reluctance in 
offering negative 
answers.
Partners’ field staffs may be more 
forthcoming.
Some partners will have their 
offices in the field, probably most 
will be headquartered in Dhaka.
Extended interviews with all 
9 partners will not be 
possible, but we should be 
able to meet those working in 
the field areas we will visit.
Partners will have 
documentation 
required by 
USAID and GOB. 
Some may have 
conducted 
analyses on their 
own that will be 
useful.
(b) other donors 
(e.g. World Bank 
LGSP)
LGSP has been the 
heavyweight player in 
local govt programming.
We will likely find some other 
donors active in the areas we 
want to visit, and will try to touch 
base with their field operations in 
those areas.
Other donors: UNDP, SDC, 
DANIDA, JICA.
World Bank 
should have much
documentation, 
other donors as 
well.
(c) and 
beneficiaries, 
such as local 
government 
units, 
UP/PS secretaries 
generally have the 
most useful 
information.
We will seek out LGU secretaries, 
officers in sectors (TBD) 
supported by SDLG activities. 
(d) locally elected 
officials,
LGUs engaged in UZP 
elections should be 
avoided (too much 
distraction)
(e) citizens, CiG members will be interviewed 
(esp. women members).
(f) and GOB? GOB officers in LGRDC 
ministry to be interviewed.
LGRDC reports 
as available.
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 92
Evaluation
Question
Assumptions/Comment
s
Field Visit Dhaka Interviews Document Review
4. Sustaina￾bility:
How sustainable 
are SDLG 
program 
activities?
Query: to what extent 
have USAID-supported 
LG programs enabled 
LGUs to avoid 
providing their own 
sustainability?
What do local elected officials 
think is worth keeping after EOP? 
What are they willing to pay for?
BUPF, MAB, UzPAB leaders 
– what do they think is worth 
keeping after EOP?
What are MLGRD views on 
this?
How prominent 
has sustainability 
been featured in 
USAID guidance 
to SDLG?
How has it been 
framed?
What measures 
could have been 
taken to enhance 
sustainability?
How far could USAID 
and SDLG have 
pushed LGUs to 
increase local revenue 
mobilization?
Given the low % local 
revenues have 
comprised of the LGU 
budget, even if fully 
collected, was it wise 
to emphasize this 
program component so 
much?
LRM will be a major focus of our 
field interviews:
(1) track record of LGUs;
(2) views of elected officials, govt 
servants, citizens on LRM;
(3) commitment at LGU level to 
emphasize LRM post-SDLG. 
SDLG surveys 
show detailed 
data on LRM, 
which will be 
useful in our field 
visits.
Note: LRM = local revenue mobilization 
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 93
Evaluation
Question
Assumptions/Comme
nts Field Visit Dhaka Interviews Document Review
5. Rele￾vance: To 
what extent are 
the project’s 
objectives still 
relevant to the 
current 
development 
circumstances in 
Bangladesh,
GOB policy on LG 
has always featured 
strong rhetorical 
support combined 
with implementation 
through patronage to 
benefit ruling party 
members while at the 
same time delivering 
some local-level 
development. This 
pattern cannot be 
expected to change 
under the present AL 
ministry. Even so, 
there is room for 
maneuver.
BUPF and MAB have been trying to 
mobilize their members to engage 
in advocating policy change at the 
macro-level. We will ask how 
responsive are local member units 
of these bodies to these desires
from above. We will also seek their 
views on SDLG efforts to build 
capacity among the member units. 
What are local views on citizen 
involvement in LGU activity? Has 
BUPF, MAB, UZPAF leaders 
in Dhaka will be asked for 
their views on training 
received from SDLG on 
advocacy and capacity 
building.
In the past, BUPF 
and MAB 
documents have 
not been hugely 
helpful. After TA 
from SDLG, this 
may have 
changed.
and will they 
provide sufficient 
guidance for 
appropriate 
programmatic 
and technical 
assistance 
decisions?
Do respondents at local level 
believe it will continue to be worth 
while to pay annual dues to BUFP 
and MAB?
Will citizen groups want to continue 
investing in the effort to have a 
voice in determining local public 
policy?
What are BUPF/MAB/UZPAF 
plans for providing guidance 
to their members in the 
future? Can they be weaned 
from reliance on USAID for 
their basic support?
Doubtful that 
documents will be 
helpful with this 
topic.
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 94
Evaluation
Question
Assumptions/Comme
nts Field Visit Dhaka Interviews Document Review
6. Cross￾cutting 
Issues:
To what extent 
were gender and 
youth effectively 
addressed by 
SDLG’s 
interventions in 
the targeted 
areas?
Our impression 
based on previous 
Bangladesh 
experience, is that 
gender equality 
objectives have 
made some real (if 
imperfect) progress 
at local level…
All field interviews, especially group 
interviews,will explore women’s 
inclusion in LGU activities. 
Additionally, women’s role in UPs, 
Ward councils, standing 
committees, and CiG groups will be 
explored. Did women benefit In 
ways different from men (e.g., 
inclusion in decision-making for the 
first time)?
SDLG staff and NGOs 
implementers will be 
interviewed on this issue.
SDLG reports
… but youth have not 
been included in any 
meaningful role. 
Both impressions will 
be empirically tested 
in our field visits
All field interviews, especially group 
interviews, will explore youth 
inclusion in LGU activities. 
Women’s role in UP, Ward councils, 
standing committees, and CiG 
groups will be explored. 
SDLG staff and NGOs 
implementers will be 
interviewed on this issue.
SDLG reports
Final Performance Evaluation of Strengthening Democratic Local Governance (SDLG) Project 95
U.S. Agency for International Development 
1300 Pennsylvania Avenue, NW 
Washington, DC 20523