9 August 2026: Expansion of Shrimp Production in Bangladesh – What Does the Satellite Data Look Like?

---

PPT slides here.

Highlights from the Q&A session:

Question: Why is shrimp farming declining in Khulna District?

Response: Shrimp farming in Khulna District is declining due to a combination of environmental, production-related, and socioeconomic factors. One of the major environmental factors is increasing salinity pressure, which can affect soil and water conditions and increase production risks. In addition, disease outbreaks, limited availability of high-quality shrimp fry, and other production-related risks can reduce shrimp productivity and profitability. Small and marginal farmers may also face limited access to capital and credit, while market price fluctuations and uncertainty can further increase the economic risks of shrimp farming. Together, these factors can increase production costs and reduce profitability.

Question: Can surface salinity vary between the dry and wet seasons?

Response: : Yes, surface salinity can vary significantly between the dry and wet seasons. In general, salinity tends to be higher during the dry season due to reduced rainfall, lower freshwater availability, increased evaporation, and greater influence of saline water. During the wet season, increased rainfall and freshwater flow usually reduce the salt concentration, leading to lower surface salinity. However, in coastal areas such as Khulna and Bagerhat, salinity is also influenced by tidal activity, freshwater discharge, evaporation, and saline-water intrusion. Therefore, seasonal variation is an important factor when interpreting surface salinity conditions.

Question:How did you distinguish shrimp farming areas from other water bodies that may have similar physical or spectral characteristics?

Response: I did not depend on spectral characteristics alone because shrimp ghers and other water bodies can have similar spectral responses. I mainly used visual interpretation of satellite imagery, together with the geometric shape, spatial arrangement, and surrounding land-use context. Shrimp ghers generally have organized and relatively regular boundaries and often occur in clustered patterns. These characteristics helped me distinguish them from many natural or other water bodies.

Question: Can salinity be measured or estimated using satellite imagery? If so, how did you assess surface salinity?

Response:Yes, satellite imagery can be used to assess and estimate the spatial and temporal variation of surface soil salinity. Several remote sensing indices and empirical approaches have been developed for this purpose. One commonly used approach is the Normalized Difference Salinity Index, or NDSI. In my study, NDSI index was used to identify salinity affected areas and non-salinity affected areas to examine how salinity conditions changed over time. However, it is important to clarify that a satellite-derived salinity index is not a direct measurement of salinity. It is an indirect spectral indicator or proxy of salinity conditions. For more precise quantitative estimation, satellite-derived indices can be calibrated and validated using field measurements such as electrical conductivity (EC). Therefore, satellite imagery provides an effective way to monitor the spatial and temporal patterns of surface salinity over large areas.

Question:Is there any relationship between surface salinity and settlement patterns? Did your study consider this relationship?

Response: There can be a relationship between surface salinity and settlement patterns, particularly in coastal agricultural areas. Increasing salinity can reduce agricultural productivity, affect freshwater availability, and create difficulties for local livelihoods. These conditions may potentially influence migration and settlement patterns. However, this relationship was not directly examined in my study. One of the main objectives of my study was to examine the relationship between surface salinity and shrimp farming. Therefore, the potential relationship between salinity and settlement patterns was recognized, but it was beyond the main scope of this research.

Question:Why did you use different study years for Khulna and Bagerhat?

Response: For the LULC analysis, I used 1991, 2000, 2011, 2020, and 2026 for Khulna District, while for Bagerhat District I used 1990, 2000, 2010, 2020, and 2026. For the surface soil salinity analysis, satellite imagery from 1991, 2000, 2011, 2020, and 2026 was used for both districts. The slight difference in the LULC years was mainly due to the availability and quality of suitable satellite imagery. It was not always possible to obtain cloud-free and sufficiently clear images for both districts in exactly the same year. Therefore, when an image from the target year was unavailable or unsuitable, an image from a nearby and representative year was selected. This provided a sufficiently long-term temporal framework for analyzing LULC and salinity-related changes. However, I also acknowledge that the difference in the LULC reference years is a limitation of the study, because the temporal intervals are not exactly identical between the two districts.

Question:Did you integrate any socioeconomic data into your analysis?

Response: No. I did not integrate socioeconomic datasets into the analysis. The study mainly depended on satellite imagery and remote sensing-derived information to investigate the long-term spatio-temporal expansion of shrimp farming, associated land-use and land-cover (LULC) changes, and surface salinity dynamics, with particular emphasis on their regional diversity across the coastal districts of Khulna and Bagerhat.

13 July 2026: Closing Bangladesh's Gender Wage Gap: Beyond Education and Skills

---

PPT slides here.

Highlights from the Q&A session:

Question: Since the unexplained component does not fully capture discrimination, is there any way to measure the amount of discrimination directly?

Response: Not with the current data. The LFS records a limited set of observable characteristics, so the unexplained component absorbs both unequal treatment and factors the survey does not measure, such as job tenure, firm characteristics, specific tasks, and hours flexibility. A purpose-built survey designed to capture these additional determinants of earnings would shrink the residual and allow the role of discrimination to be bounded more precisely. However, since no survey can measure every wage-relevant factor, decomposition methods can become more precise but never fully accurate in isolating discrimination. Measuring discrimination directly requires different designs altogether, such as correspondence or audit studies, where equivalent applications differing only by gender are compared, which could complement survey-based decomposition work in Bangladesh.

Question: Does this study argue against investing in women's education?

Response: : No; and the distinction matters. Education remains essential: it is how women enter paid work at all, and the study's own estimates show that returns to education are higher for women than for men at every level. The finding is narrower: for women already in wage employment, the gap is not caused by a shortfall in qualifications, so further education alone cannot be expected to close it. Education gets women into the room; it does not, by itself, determine how they are paid once there. The policy implication is therefore not to redirect resources away from education, but to complement it with interventions aimed at the institutions that set and enforce pay, equal-pay enforcement, pay transparency, and wage protection in informal work, which is where the evidence locates the gap.

Question: In some informal-sector work, physical capacity differs between men and women- in jute processing, for example, a task that takes a female worker twelve hours may be completed by a male worker much faster, so owners prefer hiring men. How is this captured in a gender wage gap analysis?

Response: What the question describes is productivity or 'ability' heterogeneity, which the LFS does not observe; in a decomposition framework it is therefore absorbed into the unexplained component rather than the explained one. Capturing it properly would require task- or firm-level productivity data, such as piece-rate output records or linked employer-employee data, which are difficult to obtain in a nationally representative study. It is worth discussing how ability across different industries might be proxied in future survey instruments. Two caveats apply: physical capacity is task-specific rather than general, and the decomposition question concerns pay conditional on being hired; differences in hiring itself are a separate, and also important, margin of the gap.

Question: How is potential experience calculated for workers?

Response:Following the standard Mincerian convention, potential experience is calculated as age minus years of schooling minus six, and enters the earnings equations linearly and as a quadratic. It is 'potential' rather than actual experience because the LFS does not record work histories: the measure assumes continuous participation after leaving school.

Question:How would you describe your paper in two simple lines? And could you explain the market bag example for the Oaxaca method once more?

Response: In two lines: women in Bangladesh's wage employment are at least as qualified as the men working beside them, yet they earn around 35 per cent less per month. The gap is therefore not a qualifications problem and closing it requires reforming how pay is set rather than only investing further in education. The market bag analogy works as follows: Imagine two shoppers at different tills whose bills come out different. There are only two possible reasons: either their bags contain different items, or the two tills are charging different prices for the same items. The Oaxaca-Blinder decomposition asks the same question about wages: the 'bag' is what a worker brings to the job (education, experience, occupation, location) and the 'till' is the labour market's price list for those characteristics. The method takes women's actual bag and runs it through the men's till: what would their characteristics earn if priced the way men's are? In the 2023 data, women's bags are at least as full as men's, so almost none of the gap comes from the bags; nearly the entire difference lies in the till. Women are paid less for the same contents, plus whatever the survey cannot see inside the bag, which is why the residual is interpreted carefully rather than read directly as discrimination.

5 March 2026:Assessing Implementation Progress of National Social Security Strategy (NSSS) Action Plan II (2021–2026)

---

PPT slides here.

Highlights from the Q&A session:

Question: For the scoring analysis, was there a system that considered weighted scoring, so that more important performance indicators received greater importance in the analysis?

Response: No. All performance indicators were assessed with equal weight. However, a weighted approach could be a useful refinement, particularly if future exercises want to reflect differences in the relative importance of indicators.

Question: Was the scoring subjective?

Response: Yes. The scoring involved an element of judgement, as it was based on team discussion and evidence gathered through key informant interviews (KIIs) to arrive at agreed ratings for each performance indicator.

Question: How were indicators treated when no data was available?

Response: Indicators with no available data were excluded from the scoring exercise.

Question: Would that not create an exclusion error?

Response: The unassessed indicators were documented in the report. In total, 18 out of 137 indicators were unassessed, so the overall effect on the scoring results is limited.

Question: Overall, the number of unassessed indicators is low, but for some thematic areas the share of unassessed indicators is high. What should be done in this case?

Response: This points to an information gap. For the next phase of the NSSS, these data gaps should be clearly highlighted and addressed, so that future scoring exercises can be completed without facing repeated gaps in indicator-level information.

5 January 2026: Estimating the Potential Tax Revenue from Direct and Indirect Sources in Bangladesh: Evidence from HIES 2022

---

PPT slides here.

Highlights from the Q&A session:

Question: How did you estimate wealth?

Response: Wealth is estimated by summing the reported values of all household assets, such as housing, land, businesses, financial assets, durable goods, savings, etc and subtracting outstanding loans. However, some limitations definitely exist. Like, some important asset valuations are not provided in HIES. For instance, if an individual owns multiple houses, we could only add the valuation of the house he/she is living in. Moreover, asset values are primarily based on self-reported valuations from the HIES 2022, and land values are imputed using local median prices where direct valuations are not available.

Question: It seems unlikely that the potential VAT is 16% of GDP, as the highest imposed VAT rate on any domestic good/service is 15%. What could be the reasoning for that?

Response: This issue requires more scrutiny. Can’t provide an instant answer, have to go through the estimation methodology again.

Question: Is there any international practice of e-invoicing?

Response: Yes. Many countries have adopted e-invoicing as a standard tool to improve VAT compliance and reduce tax evasion. Countries such as Chile, Brazil, Mexico, Italy, Turkey, and India require businesses to issue electronic invoices that are reported in real time or near real time to tax authorities.

26 October 2025: Assessing the Impact of Maternal and Newborn Care on Neonatal Mortality: Evidence from Bangladesh Demographic and Health Survey 2017-18

---

PPT slides here.

Highlights from the Q&A session:

Question: Did you perform any robustness checks?

Response: Not specifically. However, we also performed the regression analysis using a logit model, and the results were qualitatively similar. Apart from that, we conducted a multicollinearity test where the Variance Inflation Factor (VIF) on average was around 2, indicating no evidence of multicollinearity.

Question: Do you have any explanation for why 35% of neonatal deaths are occurring among rich households?

Response: This could be because deliveries in wealthier households are more likely to take place in hospitals or medical institutions, where reporting is more systematic. So, neonatal deaths in these households are more likely to be captured in the data.

Question: Do you think there is any issue with the methodology of the dataset that could affect the results?

Response: The survey collected information from mothers by asking questions regarding births that occurred in the three years preceding the survey, and they provided answers by recalling. Therefore, it is possible that the responses to certain questions, such as whether a clean delivery kit was used, whether instruments were sterilized, or a skilled birth attendant was present, may be affected by recall bias. Additionally, questions about child deaths may not always be accurately answered, as such experiences can be traumatic or emotionally distressing for mothers.

17 August 2025: Revisiting the EKC Hypothesis Through Green Budgeting: An Econometric Analysis of Selected Asian Economies

---

PPT slides here.

Highlights from the Q&A session:

Question: Why did you choose ecological footprint as the dependent variable instead of CO₂ emissions?

Response: Ecological footprint provides a more holistic measure of environmental degradation. It accounts not only for carbon emissions but also for resource use and ecosystem capacity, making it a stronger indicator for linking with green budgeting policies.

Question: Your results suggest the EKC hypothesis is not confirmed for most countries. What does this imply for policymakers?

Response: It implies that countries cannot rely on income growth alone to improve environmental outcomes. Instead, proactive fiscal measures—such as well-designed environmental taxes, subsidies for clean energy, and reinvestment of resource rents—are necessary to ensure sustainable development.

Question: How effective were environmental taxes in the countries you studied?

Response: The impact was heterogeneous. In Japan, environmental taxes showed a significant role in reducing ecological footprint. But in countries like Bangladesh, China, and Turkey, environmental taxes were positively correlated with environmental degradation, indicating design or implementation challenges in those fiscal policies.

Question: Why was the FMOLS method used in your study?

Response: FMOLS corrects for issues of endogeneity and serial correlation, providing more robust long-run estimates compared to traditional OLS. Given the cross-country dataset and the need for reliable inference, FMOLS was a suitable choice.

10 July 2025: Methodology of the structural gravity modelling

---

PPT slides here.

Highlights from the Q&A session:

Question: Is there any proxy to capture unobserved trade costs in the augmented gravity model? You have mentioned about taking 3-5 years lag when estimating gravity model, why does that matter?

Response: With the exporter and importer remote index, one can capture some of the unobservable trade costs in the gravity model, but this proxy can not capture all of the unobserved trade costs, so it is better to use exporter-time and importer-time fixed effects. Interval panel data should be employed in order to allow for adjustment in bilateral trade flows in response to trade policy or other changes in trade costs.

Question: How can we capture the multilateral resistance term in the gravity model?

Response: The multilateral resistance terms should be accounted for by exporter-time and importer-time fixed effects in a dynamic gravity estimation framework with panel data. It should be noted that in addition to accounting for the unobservable multilateral resistance terms, the exporter-time and importer-time fixed effects will also absorb the size variables (exporter and importer GDP) from the structural gravity model as well as all other observable and unobservable country-specific characteristics, which vary across these dimensions, including various national policies, institutions, and exchange rates.

Question: If we want to include intra-industry trade in the gravity model, how can we do that? If I don’t add this in the model, does it create any problem in the model?

Response: The inclusion of intra-national trade data in structural gravity estimations is desirable for several reasons. First, it ensures consistency with gravity theory, where consumers choose among and consume domestic as well as foreign varieties. Second, it leads to the theoretically consistent identification of the effects of bilateral trade policies. Third, it also enables to identify and estimate the effects of non-discriminatory trade policies (Heid et al., 2015). Fourth, it resolves the “distance puzzle” in trade, by measuring the effects of distance on international trade relative to the effects of distance on internal trade. Finally, it enables to capture the effects of globalization on international trade and to correct for biases in the estimation of the impact of RTAs on trade. So, it is important to consider intra-national trade in the gravity model. Importantly, intra-national trade data has to be constructed consistently as the difference between gross production value data and total exports. However, for intra-industry it should be at the sectoral level difference. Yes. If you don’t add intra-industry trade in the model, it may lead to endogeneity problem in the model.

19 June 2024: Income Taxation in Bangladesh: Facts and Parametric Estimates

---

PPT slides here.

10 April 2025: Principles of Social Insurance

---

PPT slides here.

Highlights from the Q&A session:

Question: Do you think the current budgetary arrangement for Bangladesh’s existing social protection system is sufficient and sustainable?

Response: The current budgetary arrangement is neither sufficient nor sustainable. While allocations have increased nominally, they remain inadequate relative to the population's needs and are heavily skewed towards pension payments, leaving limited resources for broader risk coverage such as unemployment, sickness, or maternity insurance.

Question: If no proof of income is required for poor individuals to enroll into Samata scheme of UPS, then how can the government identify if he is actually eligible for the program or not?

Response: The enrollment is done on the basis of self-proclamation, so if a potential beneficiary claims that they live below the poverty line, that will be the considered enough information.

Question: The presentation mentioned that Bangladesh has multiple pension schemes. How does the Universal Pension Scheme differ from traditional public service pensions?

Response: Unlike traditional schemes that are non-contributory and restricted to public employees, the Universal Pension Scheme is contributory and open to all citizens, including informal workers and expatriates. It is segmented into four schemes (Samata, Surakkha, Probash, and Progoti), each targeting specific population groups with tailored contribution rules.

Question: What portion of social protection budget in Bangladesh is allocated to social insurance?

Response: : In the fiscal year 2024–25, Bangladesh allocated approximately 33.5% of its total social protection budget to social insurance programs. The social insurance category includes three major programs:
• Pension Management
• Interest on National Savings Schemes
• Universal Pension Scheme (Samata)

6 March 2025: Benchmark Survey on Financial Inclusion in Bangladesh

---

PPT slides here.

Highlights from the Q&A session:

Question: Can the important contributions of financial inclusion, such as reducing poverty and increasing welfare, be measured using survey data?

Response: Yes, the important contributions of financial inclusion, such as reducing poverty and increasing welfare, can be measured using survey data. This survey captures key indicators like access to banking services, usage of financial products, purpose of using financial services, income and literacy levels, and improvements in household welfare.

Question: Why is the bank app utilization percentage zero for females in rural areas?

Response: The bank app utilization percentage for females in rural areas might be zero because agent banking services, which go door-to-door, are more accessible and popular among them. These services allow females in rural areas to conduct financial transactions without the need for smartphones or internet access, making digital banking apps unnecessary.

Question: If someone has access to financial services and their parents also use it, who is considered financially included?

Response: In this scenario, the person with direct access to financial inclusion would be considered financially included, as financial inclusion typically refers to an individual's ability to access and use financial services. While the parents' usage of financial inclusion services may indirectly benefit the person, but the person who has account in any financial institutions is considered to be financially included.

1 March 2025: Econometric analysis regarding the probability of a woman justifying household physical abuse based on various economic, educational, social, and circumstantial factors: using Multiple Indicator Cluster Survey (MICS) 2019 data

---

PPT slides here.

Highlights from the Q&A session:

Question: Why didn’t you include any regional dummy variables?

Response: The study aimed to focus on broader socioeconomic factors rather than regional disparities. Since variables like education, wealth index, and media exposure already capture significant differences across regions, adding regional dummies could lead to redundancy or multicollinearity. However, future studies could explore regional variations in more detail.

Question: Why didn’t you check for robustness?

Response: Given the scope of the study, the primary focus was on the logistic regression model and its key determinants. While robustness checks are valuable, they weren’t included due to time constraints and the need to keep the analysis straightforward.

Question: Why is a logistic regression model used instead of a probit model?

Response: Logistic regression is chosen mainly because it provides results in the form of odds ratios, which are easier to interpret for policy analysis. While both models produce similar results, logistic regression is computationally simpler and more widely used in applied research, making comparisons with existing studies more straightforward.

Recently held seminars:

13 January 2025: The Geopolitical dance of innovation

---

PPT slides here.

Highlights from the Q&A session:

Question: How are nations using Chinese funding despite allegations of a debt trap?

Response: Many nations, such as Sri Lanka and Pakistan, turn to Chinese funding because other financing options are either unavailable or come with less favourable terms. For these countries, China's loans often represent a critical lifeline to fund infrastructure projects and stimulate economic growth. From China's perspective, these agreements are approached as business transactions with legally recognised governments, allowing Beijing to sidestep allegations of exploitative practices by framing its role as a neutral investor. Contrary to debt-trap allegations, there are examples where Chinese funding has positively impacted recipient countries. For instance, Ethiopia's collaboration with China on the Addis AbabaDjibouti Railway significantly improved trade connectivity and logistics efficiency in the region. While these projects can create economic opportunities, the extent of their benefits often depends on how effectively nations manage the borrowed funds and the associated debt. Thus, the narrative around Chinese funding is not entirely one-sided, as there are cases of both success and controversy.

Question: What does the change in investment for R&D represent?

Response: In 1978, the U.S. government accounted for over 50% of all R&D funding in the country, but by 1996, this figure had dropped to 33.6%. This shift highlights a broader trend: during periods of heightened geopolitical tension, such as the Cold War, government investment in R&D often reaches exceptional levels to maintain competitive advantages, especially in areas like defense and technology. Over time, as geopolitical pressures ease, these investments tend to stabilise, with private sector contributions taking on a larger role in innovation and development. This change reflects the adaptive nature of funding priorities in response to the global political landscape.

Question: How does the Belt and Road Initiative (BRI) represent geopolitics?

Response: The Belt and Road Initiative (BRI) is a prime example of how geopolitics extends beyond territorial disputes to encompass economic and technological influence. By connecting over 140 nations through infrastructure, trade routes, and digital networks, the BRI enables China to expand its strategic and economic reach. It embodies geopolitics by leveraging investments and partnerships to secure influence in key regions, particularly in Asia, Africa, and Europe, while fostering economic dependency among participating nations. The initiative also highlights geopolitical competition in the digital realm through its "Digital Silk Road" component, which involves the export of Chinese technologies and infrastructure. While it provides essential development opportunities for participating nations, critics argue that it enables China to exert significant political and economic leverage, underscoring the complex interplay of power, influence, and development in modern geopolitics.

25 September 2024: Man-Made Fibre Apparel: Elevating Bangladesh’s Apparel Sector Amid LDC Graduation

---

PPT slides here.

Highlights from the Q&A session:

Question: Why is China reducing its share of cotton apparel in the global apparel market?

Response: China is now focusing on exporting MMF apparel, which constitutes more than 50 percent of its total apparel exports. This shift aligns with the growing global demand for MMF apparel, particularly in markets such as Europe, the USA, and the UK. Additionally, China is increasingly concentrating on producing and exporting electronic products.

Question: Producing MMF apparel has some environmental issues, such as carbon footprint and non-biodegradable properties. Then why should Bangladesh focus on producing MMF apparel?

Response: While both cotton and MMF apparel have environmental impacts, producing cotton requires a vast amount of water and land compared to MMF. As major export destinations are increasingly choosing MMF apparel, Bangladesh should not overly focus on environmental issues. Additionally, environmental protection measures are set by developed countries, and Bangladesh will comply with these regulations when they are established.

Question: Can Bangladesh compete with China regarding MMF apparel?

Response: China is much ahead of Bangladesh in terms of backward linkages, skilled labor force, and raw materials. To compete with China, Bangladesh needs to strengthen its backward linkages. Therefore, investment in workforce development and modern technology, expansion of research and development (R&D) initiatives, investment in man-made fiber production, and fostering innovation in recycling are much needed.

Question: How will the reduction of the import duty on MMF solve the dual concentration problem?

Response: By lowering these duties, the cost of MMF decreases, making it more competitive with natural fibers like cotton. This encourages textile manufacturers to diversify their production, reducing reliance on a single type of fiber. Additionally, lower import duties can stimulate local production of MMF, enhancing the competitiveness of domestic manufacturers.

14 October 2024: Where Does Bangladesh Stand in Terms of Achieving Gender Equality?

---

Highlights from the Q&A session:

Question: Can you elaborate on the Gender Parity Index? How are the numbers interpreted?

Response: The Gender Parity Index (GPI) is a measure of equality between men and women in key areas like education, employment, and political participation. A GPI value of 1 indicates full parity, meaning men and women have equal access. Values below 1 show women are disadvantaged

Question: Why do you think women’s unpaid contributions are not included in GDP calculations?

Response: Women’s unpaid work represents a significant contribution to the economy and should ideally be recognised. However, some debates argue that directly including it in GDP calculations could inflate the data and make it incomparable to other countries unless adopted globally. Any such inclusion needs to be done in a coordinated manner, ensuring standardisation across nations for consistency and comparability.

Question: How can the barriers preventing women from reporting harassment be overcome?

Response: Addressing social stigma is crucial through awareness campaigns that challenge victim-blaming and promote gender sensitivity. Strengthening legal framework implementation, ensuring accessible and supportive reporting mechanisms, and building trust in enforcement systems can also empower women to come forward and seek justice without fear of judgment or reprisal. There are numerous laws, regulations, and policies already in place to protect women and address harassment. However, the lack of proper implementation remains a critical issue. Ensuring the effective enforcement of these existing frameworks should be the utmost priority to create a safer and more supportive environment for women.

25 September 2024: Towards a WTO-compatible Export Support Mechanism: Implications for Bangladesh

---

PPT slides here.

Highlights from the Q&A session:

Question: Did the researchers find any examples of how already graduated LDCs support their exporters?

Response: Bangladesh holds a unique position among graduating LDCs, as no other graduated LDC has experienced export-led economic growth on the same scale. Consequently, the researchers could not identify any such examples, as almost all graduated LDCs are smaller economies with less emphasis on export.

Question: If Bangladesh were included in Annex VII (b) of LDCs, would it be possible for Bangladesh to be competitive with all kinds of products it exports?

Response: Even if Bangladesh were included in Annex VII (b), it would not be able to provide export supports for its major export items, as 28 Bangladeshi export items (in jute and RMG) have already reached the export competitiveness threshold. However, Bangladesh could provide support for other products that have the potential to be export competitive but have not achieved that status due to various reasons.

Question: In the examples of how other countries support exporters, there were instances that appeared to be WTO-compatible. How can some countries continue providing support even though it seems that such support is not WTO-compatible?

Response: While some export support measures may appear to violate the WTO’s ASCM, not all these measures are necessarily challenged at the WTO's DSB. There are several reasons for this. Countries often weigh various factors before deciding to pursue a case. They may choose not to challenge certain subsidies for diplomatic reasons or to maintain positive trade relations. A country is more likely to initiate a dispute only when subsidies cause substantial harm to its domestic industries or international trade interests. If the impact of a subsidy is minimal, the potential benefits of challenging it may not justify the effort. Moreover, WTO disputes can be costly and time-consuming. Smaller or developing countries may lack the financial resources or legal expertise to pursue a case, even if they have legitimate grounds to do so. In many instances, countries prefer to resolve trade disputes through bilateral negotiations rather than resorting to the formal DSB process.

Question: Can the researchers provide any insight into whether and how much success the recommended policies would bring for exporters? Which recommendations are expected to be the most successful for exporters?

Response: Estimating the extent of success the recommendations would bring was beyond the capacity of the current study.

Question: As supports like cash subsidies will need to be stopped after 2029, is the government currently reducing these supports?

Response: Yes, the government has been gradually reducing cash incentives over the past year to ensure compatibility with WTO regulations. However, another approach could be to make the most of the remaining time before 2029 by effectively utilising the policy space to strengthen the foundation of certain industries.

Comments and feedback:

• Given the context that there already exists an anti-export bias in Bangladesh due to high tariff rates protecting domestic import-competing industries, government support aimed at making these industries export-competitive may further worsen the existing anti-export bias.

• The simplification of the Duty Drawback Scheme, as mentioned in Bangladesh's National Tariff Policy (NTP), can be considered.

• Since there is still room to support exporters in WTO-incompatible ways until 2029, considering due restraint period, Bangladesh can use this time to strengthen the industrial base in key sectors to cushion the post-graduation shock from the loss of some export supports. For example, the country could shift export support from cotton-based apparel to man-made fibre (MMF)-based apparel, aligning with the changing preferences of RMG buyers. Similarly, support can be extended to the leather industry to make it environmentally sustainable, making effective use of the remaining period.
Export support for agricultural goods can be provided for a longer period even after graduation through the Aggregate Measurement of Support (AMS) or by remaining on the list of Net Food-Importing Developing Countries (NFIDC). Therefore, it would not be wise to lift existing supports now. Instead, they should be continued until Bangladesh's AMS threshold crosses the de minimis level or it exits the NFIDC list.

• The presentation recommended introducing regional support programmes to incentivize the decentralisation of industrial investment in Bangladesh. However, there is already a provision in place that encourages investment in disadvantaged regions.

18 September 2024: Bangladesh's Pharmaceutical Industry: Preparing for LDC Graduation and Assessing the Impact on Exports Post-Graduation

---

PPT slides here.

Highlights from the Q&A session:

Question: During 1980s the pharmaceutical industry was protected through limiting the activities of MNCs. After LDC graduation, we can’t impose such measures and will have to open the market. What impact will it have on the industry if we allow competition?

Response: While increased from MNCs may impact local firms’ profitability, it will have a positive impact on consumer. Consumers will be able to purchase medicine at a low cost. Also, the efficacy of medicine may be improved by allowing MNCs in this industry.

Question: How the incorporating parallel importation may influence price of on-patent medicine?

Response: The parallel importation flexibility enables importers to procure medicines from markets where they are available at lower prices. By introducing this option, manufacturers are compelled to consider potential arbitrage opportunities when setting high prices for patented medicines.

Question: How the adoption of Bangladesh Patent Act 2023 prepares the industry for LDC graduation?

Response: Before the adoption of the Patent Acts in 2022 and 2023, the patent system was regulated by the 1911 Patent Act. This older legislation addressed novelty in a limited manner, allowing for practices such as ever-greening. Additionally, it did not comply with the TRIPS Agreement. As a TRIPS signatory, Bangladesh was required to implement a new patent regime, and the recently adopted Patent Acts have effectively addressed these shortcomings.

Question: How would the new patent act resolve the ever-greening issues?

Response: The new patent act defines novelty in a rigorous manner. A marginal improvement from previous patented drugs will be eligible for patented.

9 September 2024: Extending Social Protection Floor Coverage- An Assessment Based National Dialogue (ABND) Approach

---

PPT slides here.

Highlights from the Q&A session:

Question: In the dialogues, it is said programmes will be finalized ‘on priority basis’. What considerations will be given priority to regarding this?

Response: The cost of implementation and immediate return/outcome of the process are two of the factors that will be prioritized, which demands of the stakeholders and immediate need of the population in discussion will also be taken into consideration.

Question: What does extending the social protection floor or coverage look like in ABND?

Response: It can be horizontal extension, where people who were not under protection before are brought under protection, or it can be vertical extension, where the level of benefit for the people already receiving social protection is increased. The basic notion is to close down the gap between the nationally defined social protection floor and actual coverage scenario in practice.

Question: Most of the stakeholders of the national dialogues seem to be from the supply side of social protection, how will the demand side be represented?

Response: The civil society, consisting of human rights activists, academicians, and journalists will be included as stakeholders, who can represent the needs of the demand side.

Question: In the life-cycle approach of social protection coverage, how will minority groups be covered?

Response: Since life-cycle stages approach categorizes beneficiaries based on age groups, if properly implemented, minority groups will be automatically covered under this approach as they also belong to one age group or another. However, Bangladesh currently addresses the needs of minority groups under the needs of covariate shocks.

Question: How is the assessment of gaps in current social protection coverage actually made?

Response: Primarily through desk research with the help of the tool called ‘Assessment Matrix’. It helps to gather relevant information in an organized way that makes it easier to identify where the gaps in policy or implementation lie.

2 September 2024: Structural Transformation in Asia-Pacific LDCs in a Global Context: A Comparative Analysis

---

PPT slides here.

Highlights from the Q&A session:

Question: Why is the services' share of GDP not incorporated as a measure of structural transformation?

Response: In most countries, the services sectors are not developed and less progressive. Moreover, productivity growth in services is relatively lower in LDCs compared to many other countries. Although services are significantly important globally, this trend is not necessarily a good thing. It is much easier to trade physical items which tend to have higher returns to scale.

Question: Why is employment share not selected as a measure of structural transformation?

Response: The paper uses four of the most widely adopted measures of structural transformation, one of which, based on existing literature, emphasises output share rather than employment share. Moreover, data on employment by sector are also not uniformly available for all countries. There is much lower consistency in the construction of employment data across a large set of countries, which might lead to comparability issues.

Question: How urbanisation is linked to structural transformation?

Response: Urbanisation dynamics worked as a vehicle for economic transformation. Moreover, the relationship between the level of urbanisation and per capita income is positive. Further, productive cities are considered an engine of economic growth where high-value-added crucial sectors have replaced low-value-added agricultural sectors. Such a transition can be linked to defining structural transformation.

Question: What is the reason for including the square of GDP per capita?

Response: The paper uses GDP per capita and its squared term to address potential non-linearity issues. The relationship between the dependent and independent variable might not always be linear (straight line), but rather it might be curve-shaped. Therefore, to control this issue the paper takes the square of GDP PC.

Question: Why is random effect estimation instead of fixed effect estimation conducted for empirical purposes?

Response: For empirical analysis and to focus on our objective, the model includes country dummy variables along with other variables in the model. Fixed effect model controls for all time-invariant characteristics of all entities. For each group, the dummy variable takes 0 or 1. In this case, fixed effect implies within transformation and that dummy gets omitted. Therefore, the estimation procedure employed panel random effect method.

Comments and feedback:

Instead of panel random effect estimation procedure, there are other econometric approaches such as difference in difference (DID), least square dummy variable (LSDG) along with year fixed and regional fixed effect might be useful to make the empirical analysis more robust. Analysis of descriptive statistics would be more appropriate to understand and make valid comparisons or differences among country groups. In addition to comparing AP LDCs with global and developing countries, the analysis should also include comparisons with other least-developed countries (LDCs).

21August 2024: Needs Assessment Study on the Necessity of More Off-dock/Dry Ports

---

PPT slides here.

Highlights from the Q&A session:

Question: Here, you prioritise rail as the mode of transportation over waterways and roads. So, why aren't we improving the waterways and roads?

Response: The waterway between Dhaka and Chittagong is not conducive for adequate container-carrying vessels. For example, the channel can accommodate ships with a maximum draft of 4.5 meters, which is not efficient for carrying a sufficient number of containers. On highways, a lorry can carry a maximum of 3 TEUs and often faces traffic congestion. However, a rail can carry more than 18 TEUs at a time without encountering traffic jams.

Question: Despite having an Inland Container Depot (ICD) in Kamalapur, why are you suggesting the establishment of additional dry ports in Gazipur and Narayanganj?

Response: A significant portion of Ready-Made Garment (RMG) industries are located in Gazipur and Narayanganj, and RMG contributes a substantial share of exports. Establishing additional dry ports closer to Kamalapur could exacerbate traffic congestion in the area. Therefore, setting up dry ports in Gazipur and Narayanganj is more strategic and efficient.

Question: What are the major policy conflicts regarding the zoning laws?

Response: Currently, there are two off-dock policy documents in Bangladesh: one prepared by the National Board of Revenue (NBR) and another by the Chittagong Port Authority. The NBR recommends that new off-docks be established within 20 kilometers of the Chittagong port, while the Chittagong Port Authority suggests that they should be located beyond 20 kilometers from the port. This difference in recommendations is a significant policy conflict.

Comments and feedback:

• It is highly recommended to prioritise rail as the primary mode of transportation for cargo handling. However, to address potential future scenarios where the railway could be damaged by disasters, it is advisable to develop an emergency alternative transport solution.
• It is recommended to integrate 5G technology into port management. Given that many off docks in Bangladesh are currently outdated, it is advisable to consider updating with proper modern equipments with details.