1. Chandrasekhar Subramanyam: Another Index used to measure Diversity is Gini Index. Did you try Gini Index and compared the results?

Subhankar Mukherjee: Yes, we did. But Gini doesn’t allow decomposition of the diversification index from food to non-food group, which we perform.

2. Chandrasekhar Subramanyam: Can you clarify what is j = 1,2 in religion?

Subhankar Mukherjee: Base is Hindu. Two other minority religious groups.

3. Gyandeep Hazarika: Madam, do you think that somehow, at present we have achieved a milestone of financial inclusion? if yes then what would be the next milestone to achieve after financial inclusion! Making them financially resilient? or Educating them towards various financial products?

Manisha Chakrabarty: Education and enabling people to use digital tools and enhancing financial literacy and having programs with wider reach would be helpful. As Prof. Gupta indicated, India is still lagging in financial inclusion. There is already some work ongoing on financial literacy and this should continue

4. Ashim Kapoor: If they dropped invalid obs, how did they have a balanced panel?

Manisha Chakrabarty: Invalid observations are merely missing and other records where full data is not available. We only remove those, we ensure the balanced panel includes all households for which data is available for the entire study period.

5. Chandrasekhar Subramanyam: What is the difference between the first paper and this?

Arpit Gupta: The paper I mentioned by Agarwal et al did not have much to say on consumption, it predominantly focused on lending implications and data from bank accounts.

6. Sunita Sharama: Besides PMJDY there can be other factors responsible for diversification in food & non-food expenditure

Subhankar Mukherjee: This was already addressed but we agree there are other factors as well. First we curtail our dataset to 2016 to avoid other macroeconomic shocks but even within this period there could have been other shocks that we could have controlled for. This is why we ran the regression multiple times with different variables. For example we had the macroeconomic trend variable and the time invariant variable. These should be able to capture these factors. The other suggestion by Arpit on using causal inference would also be useful.

Use of CMIE Product implies acceptance of the usage agreement & privacy policy   ♦   FAQs   ♦  Diagnosis