The concept of graduation has gained significance in anti-poverty programming and policies, catching the attention of policy-makers and development practitioners. Non-governmental Organisation (NGO) and donor-led and governmental graduation programmes have proliferated over recent years to serve different contexts. Graduation is particularly relevant to contexts involving exceptional crisis and risk, such as post-disaster, conflict, and in refugee camps. The approach was pioneered in 2002 by the Bangladesh NGO, BRAC, with the Targeting the Ultra-poor (TUP) programme.

Looking at the results of different impact evaluations from a gendered poverty knowledge framework, I argue that although BRAC’s TUP represents a comprehensive package to poverty alleviation that focuses on multiple aspects of poverty, the programme presents some challenges to the targeted women that are comparable to the challenges experienced by women beneficiaries of microcredit and conditional cash transfer (CCT) programmes.


BRAC’s Targeting the Ultra-Poor programme:

i. Origins 

BRAC’s TUP has its origin in an acknowledgment that microcredit programmes rarely reach the poorest populations, being a better entry point for moderately poor than for very vulnerable people [1]. BRAC initiated an assessment of the constraints faced by the ultra-poor in the Bangladeshi context and found that the ultra-poor are often food insecure; excluded from government safety nets; vulnerable to external shocks; live in hard-to-reach areas; underserved by markets; lack community acceptance; and are predominantly from a female-headed household [2].


ii. Design

The programme targets ultra-poor women and delivers a support package that includes:

  • Transfer of assets and cash,
  • access to savings and credit facilities, and
  • training during a time-bound period of 24 months.


Following receipt of the support package, participants are presumed to graduate out of extreme poverty into a life of self-reliance [3]. Potential beneficiaries are identified through a participatory community wealth ranking, where after programme staff pay door-to-door visits to establish whether households qualify based on at least three of five inclusion criteria:

  1. Households must not own land (or only a very small area) or
  2. they must not own any productive asset,
  3. no working males,
  4. no school-aged children can be enrolled in school, and/or
  5. the family must rely solely on female work or begging [4][5].


By the end of the programme, households are considered as ‘graduated’ when they meet the majority of pre-defined indicators in the areas of:

  • Food security: A minimum of 2 meals per day.
  • Economic resilience: Diversified income sources, doubled the value of assets, and improved home conditions.
  • Improved health: Use of sanitary latrine and clean drinking water.
  • Social inclusion and empowerment: Engage in household decision-making and participate in community events.
  • A positive change in behaviour: Children attend school, no child marriage, and use of family planning [6].

This way of measuring programme achievement goes beyond the conventional money-metric method of calculating wellbeing but favours multidimensional aspects of poverty and vulnerability. Moreover, several of the aforementioned indicators suggest that women’s empowerment is an important aim of the programme.


iii. Impacts:

1. Early quantitative studies have shown evidence of short- and medium-term positive effects, including:

  • increased income and ownership of productive and non-productive assets,
  • increased food and non-food consumption, and
  • a favourable shift in ownership of assets and hours spent on self-employment [7] [see statistics in note].


2. Employment and income: Risk and investment

In 2011, 7 years after baseline, the programme’s effect on employment and income started to slow down [8]. Especially among women that were beggars, maids and day labourers, and who presented entrepreneurial possibilities on short- and medium-term, had returned to their original occupations. In the long-term, the effect on income was only greater for those who were day labourers and already entrepreneurs at the onset of the programme.

The long-term fading of entrepreneurial possibilities indicates that BRAC’s TUP has similar limitations to those found in microcredit programmes, namely the qualitative difference between growth-oriented entrepreneurs and survival entrepreneurs [9]. The former is motivated and has income enough to both satisfy household needs and invest in self-employment. The latter operates in a situation of constant insecurity and risk of further destitution, thus avoiding investment risks. This slowdown in income levels may suggest that the families might not be able to sustain graduation after the programme ends.


3. Gender equality

Qualitative and mixed methods studies assessing the impacts on gender equality have found mixed outcomes [10] [11]:

  • Economic: Women’s access to productive assets and additional income generated by the owned asset allows them to expand their livelihood set.
  • Empowerment: Women tend to perceive the programme very positively, emphasising subjective impacts such as gained confidence, increased social capital, increased feeling of social status in the household and community. 
  • Household decision-making: Evidence that these improvements effectively translated into more decision-making power within the household or at community level is limited. Although the women control the transferred asset, new investments are generally owned by men, especially agricultural and non-agricultural productive assets, durables and land. Likewise, there was evidence that the programme decreases women’s decisions over thier own income, purchases for themselves, and household budgeting.
  • Work: The studies reveal the programme shifts women’s work from outside to inside the home since the asset is managed in the home sphere, further reducing their mobility. As women tend to spend more time managing livestock, it becomes an additional burden on top of existing domestic and income-earning activities.


Implications: A burden of responsibilities and obligations 

It is often women who bear disproportionate responsibility for the (often) unpaid domestic and care work. Targeting women may lead to a feminisation of obligations [12]. As long as policies and programmes do not offer support that allows families to balance work with parenting/domestic responsibilities and encourage greater participation and burden-sharing by men, women risk bearing the burden of additional responsibilities and obligations embedded in programme requirements.

The same has been suggested in relation to conditional cash transfers (CCT). While analysing the design of Mexico’s CCT programme, Molyneux [13] observed that the programme does not have women’s needs as an explicit aim, constraining them to their traditional roles as mother and carers without consideration for their breadwinning responsibilities.   


TUP shortcomings

As argued, the underlying assumptions of BRAC’s graduation programme represent a multidimensional perspective on poverty and likewise a multifaceted response. While impact evaluations show some positive effects, such as increased income levels, food security and ownership and control of assets, along with a self-perceived feeling of empowerment among targeted women, the programme fails to impact a number of important variables:

  1. Entrepreneurial possibilities are high in the short- and medium-term but erode in the long run, especially for women that are more deprived and vulnerable at the start of the programme.
  2. Women’s work and responsibility increase as a result of added responsibility of tending to the assets on top of existing domestic and childcare responsibilities.
  3. Studies show that the programme may induce a decrease in women’s decision-making power within the household.


Designing gender-sensitive social protection

The targeted approach of TUP suggests that the programme has important gender concerns, as income is a less robust indicator for women’s poverty compared to dimensions such as ownership and access to land and assets [14]. This is also implicit in the choice to target female-headed households, who have to maintain the household alone, thus facing higher risks of material and time poverty than those who can rely on another earner in the household [15].

Targeting women is only positive when there is an explicit intention of improving women’s asset ownership and their ability to make strategic life choices [16]. Otherwise, there is a risk of reinforcing women’s subordination as responsible for the private sphere, making them instrumental in the achievement of economic and social objectives [17]. 



To become more gender-sensitive, BRAC’s graduation programme should cater to the disadvantages, risks and responsibilities of women to a greater degree. This includes:

  • Addressing women’s ability to tackle and overcome cultural restrictions in order to increase women’s mobility in spaces outside the household; 
  • increasing their voices and bargaining power within the household;
  • reducing women’s domestic workloads; and
  • providing and advocating for better access to care services for children to reduce the domestic responsibilities of women.



[1] [3] Hashemi, S.M. & Umaira, W. (2011). New Pathways for the Poorest: the Graduation Model from BRAC, CSP Research Report 10, Brighton: IDS. Accessible:

[2][6] Whitehead, L. (2016). The Ultra Poor Graduation Approach - Lauren Whitehead (UNRISD Seminar), Youtube Video, August 9, 2016. Accessible:

[4] BRAC. (2013). An end in sight for ultra-poverty Scaling up BRAC’s graduation model for the poorest, Briefing Note #1: Ending Extreme Poverty, BRAC November 2013. Accessible:

[5] Matin, I., Sulaiman, M., & Rabbani, M. (2008). "Crafting a graduation pathway for the ultra poor: Lessons and evidence from a BRAC programme", RED Working Paper, BRAC.

[7] For instance, after two years, the women would devote more time to self-employment (increase of 113%) and less in wage employment (decrease of 13%). Also, the annual average income increased by 36%, and non-food consumption per person and food items increased by 17% and 6 %, respectively. Four years after the beginning of the programme, income continued to increase, which allowed beneficiaries to expand productive assets and buy land, increasing renting by 118% and landowning by 38% in relation to the baseline.

Bandiera, O., Burgess, R., Das, N., Gulesci, S., Rasul, I., & Sulaiman, M. (2013). Can basic entrepreneurship transform the economic lives of the poor? IZA, Discussion Paper No. 7386. Accessible:

[8] Misha, F., Raza, W., Ara, J., & Van de Poel, E. (2014). How far does a big push really push?, ISS Working Paper Series/General Series, 594, 1-56. Accessible:

[9] Berner, E., Gomez, G., & Knorringa, P. (2012). "‘Helping a large number of people become a little less poor’: The logic of survival entrepreneurs", The European Journal of Development Research, 24(3), 382-396. Accessible:

[10] Holmes, R., Mannan, F., Dhali, H. H., & Parveen, S. (2010). Gendered risks, poverty and vulnerability in Bangladesh. Case study of the challenging the frontiers of poverty reduction programme (CFPR), Specially Targeted Ultra Poor II (STUP II), London: ODI.

[11] Das, N., Yasmin, R., Ara, J., Kamruzzaman, M., Davis, P., Behrman, J. & Quisumbing, A. (2013). How do intrahousehold dynamics change when assets are transferred to women? Evidence from BRAC’s Challenging the Frontiers of Poverty Reduction—Targeting the Ultra Poor program in Bangladesh. Accessible:

[12] [13] Chant, S. (2003). The ‘engendering’of poverty analysis in developing regions: progress since the United Nations Decade For Women, and priorities for the future. Accessible:

[14] Molyneux, M. (2006). Mothers at the Service of the New Poverty Agenda: Progresa/Oportunidades, Mexico’s Conditional Cash Transfer, Social Policy and Administration 40 (4): 425–449. Accessible:

[15] Grown, C. (2014). Missing women: gender and extreme poverty debate. USAID, April 30, 2014. Accessible:

[16] Kabeer, N. (1999). Resources, agency, achievements: Reflections on the measurement of women’s empowerment, Development and change, 30(3), 435-464. Accessible:

[17] Chhachhi, A., & Truong, T. D. (2009). Gender, poverty and social justice, ISS Working Paper Series/General Series, 482, 1-28. Accessible: