Information related to household finances in Bangladesh is currently collected through the Household Income and Expenditure Survey (HIES). The HIES is a repeated cross-sectional household survey that is conducted in every 5-6 years to collect detailed information on household income, expenditure and other demographic and socioeconomic variables. The HIES was conducted in 2010 to collect data from 12,240 households (BBS, 2012). The 2016 round of HIES surveyed 46,076 households and the data was expected to be released in March 2019.
The HIES is considered as the most important household survey as it provides detailed information on household income, expenditure and consumption to determine the poverty profile, standard of living and nutritional situation in the country. The HIES is also employed to identify the representative consumption basket, required to construct the Consumer Price Index (CPI) in the country. In addition, the HIES provides detailed information on health, education, socio-economic characteristics, disability, migration, remittances, microcredit, disaster management practices and the Social Safety Net Programme (SSNP) coverage of the nation. Representative household level data can also indicate the regional disparity in economic status and the potential reasons for such outcomes. Thus the HIES provides relevant data for monitoring the poverty situation, implementation of important modules that are essential to monitor household welfare. However, it misses some important information related to household welfare. For example, it does not contain information on childcare, tax payment, financial stress, training, family relation and skills and abilities. Some of the modules like income, consumption (including details of energy consumption) and employment also require extensions and modifications to keep pace with the progress the country is making.
Against this background, the Bangladesh Bureau of Statistics (BBS) may conduct a new survey - The Household Income, Expenditure and Labour Dynamics in Bangladesh (BHIELD) Survey. The survey can be conducted annually applying the panel structure to collect a wide range of information to provide more frequent and accurate information about poverty, inequality, inflation and other important indicators, including those related to Sustainable Development Goals (SDGs). Availability of information would allow the government to evaluate the effectiveness of its policies and programmes and provide an incentive to follow evidence-based policy measures. It would also be helpful for the government in routinely reporting the country's progress, including the progress of SDG indicators, to the countrymen, development partners and international media.
On the other hand, having a wide range of data with an annual frequency will help the policymakers devise timely policies based on real-time information from the grassroots. The BHIELD may additionally allow and invite quality research that can bring benefit to the country, directly through receiving policy suggestions and indirectly through promoting research in the country. Overall, a high-quality household panel data would put Bangladesh ahead of other developing countries in the evidence-based policy making.
Panel (or longitudinal) data blends the inter-individual differences and intra-individual dynamics, which provides several advantages over cross-sectional or time-series data as it allows to study the importance of lags in behaviour (or the result of decision making) which is important as many economic policies show the signs of effect (or the effects become visible) only after some time has passed.
Panel data also provides more accurate inference of model parameters through more degrees of freedom and more sample variability than cross-sectional data and facilitates greater capacity for capturing the complexity of human behaviour than a single cross-section or time series data, including constructing and testing more complicated behavioural hypotheses, controlling for the impact of omitted variables as it is common in empirical work, uncovering dynamic relationships like the annual movement of the poverty head count ratio (which can validate the effectiveness of government programmes) and providing micro foundations for aggregate data analysis.
Moreover, panel data simplifies computation and inference in the analysis of non-stationary time series, the presence of measurement errors, and employing dynamic Tobit models.
Due to such advantages, a large number of government programmes and policies could be evaluated using survey data when detailed data is collected more frequently and the survey has the panel structure. In particular, sometimes government programmes are rolled out frequently such as they do not have time to collect the baseline or the follow-up information. Having a panel household survey may enable researchers and policy makers to examine the effects of many such interventions and would allow policy feedback to the government on a continuous basis.
In many developed and developing countries, there are a number of important household panel surveys. These include German Socio-Economic Panel (SOEP); Household, Income and Labour Dynamics in Australia Survey (HILDA); British Household Panel Survey (BHPS); US Panel Study of Income Dynamics (PSID); Pakistan Rural Household Panel Survey (PRHPS); Indonesian Family Life Survey (IFLS3); and Vietnam Household Living Standards Surveys (VHLSS).
Despite a few shortcomings, some panel household surveys from developing countries, like Indonesian Family Life Survey (IFLS) and Vietnam Household Living Standards Survey (VHLSS), are considered very useful. For instance, the IFLS was employed in investigating the merits of ability in developing and developed countries' multidimensional poverty dynamics; the impact of health card programme on access to reproductive health services; long-term health effects of Ramadan fasting during pregnancy on next generation; the value of vocational education; individual income inequality and its drivers; and household structures and savings. The IFLS also investigates the impact of cooking with firewood on respiratory health; gender discrimination in earnings; the role of spousal income in the wife's happiness; health status of women in major cities; effects of local governance on contemporary development; family hardship and the growth of micro and small firms; poverty and mental health; complementary policies to increase poor people's access to higher education; how education breaks the cycle of poverty; religion and cooperative attitudes; demand for food; natural disasters and vulnerability and long-term economic growth and the standard of living.
Melbourne Institute of Applied Economic and Social Research of the University of Melbourne, which conducts the HILDA survey, may lend support for the BHIELD for two reasons. The HILDA is internationally recognised as a great panel survey, and a number of researchers who are familiar with the HIES have also employed the HILDA for research and, in the preliminary stage of the development of BHIELD, they will be able to help in developing and transferring the technical know-how they learnt in Australia.
The HIES was not conducted on a regular basis in Bangladesh. Starting as Household Expenditure Survey (HES) in 1973-74, an income module was included in the 2000 wave, when it was renamed as Household Income and Expenditure Survey (HIES) (BBS, 2012, 2016). After that, the HIES was conducted in 2005, 2010 and 2016. One reason for the large interval between the rounds of HIES is the reliance on donors for financing the survey. Thus government financing may allow timely and regular implementation of the proposed BHIELD survey.
In case, foreign assistance is preferred for other reasons, some development partners can be interested in funding the BHIELD. These are: United Nations, which is interested in monitoring SDGs; World Bank, which is currently funding HIES, and Australian Government, particularly if Melbourne Institute (which administers the HILDA survey) is involved with the proposed project.
The frequency of some panel data like HILDA is annual. The HIES currently collects data in every 5-6 years. The survey, in its panel format, can be made annual/bi-annual. Again, important panel surveys contain a great number of modules. Depending on the relevancy and the availability of funding, the frequency of the modules of the proposed survey can be organised accordingly.
However, the use of up-to-date information and communication technology (ICT), together with the recently developed National ID system in Bangladesh is expected to track households relatively easily and put downward pressure on the cost of data collection. It might also be possible to integrate some other surveys into this project. In such a case, funds from such projects can be used in the proposed survey. As a result, the marginal cost of implementing the project is expected to be low and would not be of any concern.
Based on the scenario presented above, we expect Bangladesh Government to come forward to initiate a new household panel survey to collect information on household, income, expenditure and labour dynamics in the country. To reduce reliance on the donors, the survey can be financed by the government. However, in conducting the survey, the BBS can seek technical support from the Melbourne Institute of the University of Melbourne to ensure the quality of the survey. The new survey can be conducted annually to provide more frequent and accurate information about poverty, inequality, inflation and other important indicators, including those related to SDGs. It will allow the government to evaluate the effectiveness of its policies and programmes to improve policy measures based on evidence. It will thus enable the government to report the country's progress, including the progress of SDG indicators, to the countrymen, development partners and international media.
Dr Syed Abul Hasan is Senior Lecturer at School of Economics and Finance, Massey University, New Zealand and a former Senior Assistant Chief at the General Economics Division, Bangladesh Planning Commission.
S.A.Hasan@massey.ac.nz