Repository for BD Statistics (BBS) quality/credibility

[Nilgiri]

Do you think UNDP made a slide to joke on PDF. No they dont.

Oh so UNDP itself made this powerpoint thing now LOL. Clearly you don't understand how non-original content hosting works (Please check the whole powerpoint if you want to see if and where UNDP states its authorship, or anyone else does for that matter).

I will give you a cpl "not typical LDC guy" points if you are able to find through your frantic googling what the "NASA" watermark on top right is .

With your logic Bangladesh should have in the lower of the list but you just are 1 point ahead than us.

a) I am not talking about statistical raw (theoretical) capacity, I am talking specifically about quality and credibility.
b) Check more recent years at world bank site even on this raw theoretical capacity, specifically the trends BD is seeing now...you are doing the opposite of improving, esp in "methodology" component . Why is that?

Confessed by Govt itself and they are working on it.

Yet refusing to accept the ESCAP assessment (which is simply as easy as ticking a box and admitting you have a deep set problem and accepting the policy recommendations and subscribing to a timeline to address it with further auditing to prove it). Its only been 6 years now....yet this simple thing not done.. Tick friggin tock.

But obviously BD has nothing to hide at all This is btw a huge reason why BD does not progress on corruption and institutional indices...and stays in the lower accounting standard of the IMF (GDDS as opposed to SDDS)...which leads in turn to such strange things as Bangladesh still not being listed in say The Economist magazine weekly appendix of major world economies (even though The Economist lists Pakistan and even covered an article of BD surpassing Pakistan in per capita nominal GDP due to recent Pak pop census). Something is really lacking in BD then, because Pakistan is also in same GDDS category as BD in IMF.

Congratzz btw. Bangladeshi politicians need to learn this asap how you did it!

Thanks. BD should look to create and harness dedicated set of ministers purely on their merit with addressing business needs in concert with specific World Bank input/standards. Indian federal and state govt (esp in MH given bombay is one of two cities surveyed) put harsh deadline on various ministers to implement everything needed 1 month before the world bank submission deadline so that there was enough time for the World Bank enterprise surveying to flow back into its analysis. There is also good follow up process now using this iteration results for further improvement in the subcomponents we are still doing badly.

Competitive federalism (state vs state competition) is something starting big time in India, maybe BD needs to look into how it can also promote it (i.e not just be reliant on Dhaka, but have major hub cities and their districts compete more with each other in good way). I dont think 170 million, long term 200 million people can rely on just Dhaka and CTG....you need like 6+ cities and their major districts all competing and experimenting with policy....i.e some solid decentralisation. But of course the institutional credibility (courts, bureaucracy, finance conduits etc) needs to be addressed in systematic way with specific (transparently + 3rd party) measurable goals before this can really help....otherwise maybe more noise than help is created. I simply not sure what the incentive for BAL/BNP is to do this, BD people need to demand much much more, they cant rely on just 1 or two political parties based on emotions over delivery....because neither will seek to rock the bureaucracy boat too much given guaranteed short and mid term buffers from RMG etc.

To get economic diversification and broad base crediblity that it requires, BD needs to be churning through major reform pre-emptively right now (rather than token here and there stuff)...in preparation for graduating from LDC status in 2021 (if indeed it starts the process in 2018 which is very likely) and losing major tariff/quota advantages in many of its biggest clients in say Europe....but I am seeing mostly lackadaisical approach to it from BD....the BAL still cares more about whats good for its politics than the country economics...maybe that specific ratio is better than BNP but its still pretty bad. BD people should stop believing that their aspirations at this stage can be met by just 2 identity politics parties....again this feeds back into having more federal decentralised environment. Political "stability" was ok when you have 1 ticket (RMG) model that needs longer periods of known policymakers....to get more diverse economy, you need more diverse set of politics in play and more transparent way to harness the winning ones and let the losing ones atrophy. Even in developed economy a major traditional behemoth in say energy (a parallel to say BAL/BNP in RMG driven economy) is no good for competing/capturing internet/social media market and myriad of other new tech sectors (a parallel to more diverse future BD economy). But in end it depends on BD people to create new paradigms and shift to better system of politics and bureaucracy....even if it means losing some gains on paper (they only exist on paper after all, the added honesty and transparency is well worth it in long run).

 

[24 Hours]

What because you say so? Corruption Perception Index and Cato Freedom Index are also not based on any hard facts or evidence? It is wrong to say BD is stagnant there and Pakistan and India are better off (and hence more credible to begin with) and improving too? I am making it up when ESCAP assessment was rejected by BD govt?

To prove you are not a hypocrite, take me through your hard facts and evidence regarding the BBS claims on eid....start there, something simple as counting cattle. Help out your compatriot, he hit the wall hard, maybe you can climb it and keep running?:

https://defence.pk/pdf/threads/bang...-consumption-soars.525749/page-2#post-9985909

Lets see you put action to your words instead of resorting to ad-hominem. It is this sole thread I am going to hew against the BD hordes in relation to their country bureaucracy credibility.

Your constantly compare Bangladesh to North Korea and is thus untrustworthy. This is a huge exaggeration to go from BD to NK. I can't even find NK on Cato's report. Though the results should be obvious. I'll give Cato the benefit of the doubt though, it's more reliable than a CPI score.

What do you exactly want me to address in that thread? In regards to the protein consumption what is wrong with it? Increase protein consumption from 2010 over the past six years. I am not aware of India's performance in regards to nutritional development in recent years so I cannot make judgement on that and make comparisons to India.

 

[Nilgiri]

Your constantly compare Bangladesh to North Korea and is thus untrustworthy.

Corruption perception Index.

This is a huge exaggeration to go from BD to NK. I can't even find NK on Cato's report. Though the results should be obvious.

Assuming North Korea to be the worst rock bottom. How far away is BD rank wise?

I'll give Cato the benefit of the doubt though, it's more reliable than a CPI score.

How so? They aren't even really measuring the same things.

What do you exactly want me to address in that thread? In regards to the protein consumption what is wrong with it? Increase protein consumption from 2010 over the past six years. I am not aware of India's performance in regards to nutritional development in recent years so I cannot make judgement on that and make comparisons to India.

The so called BBS statistics that claim massive increase in beef production from domestic herds. BBS claims that a 50%+ yearly herd slaughter rate is both realised and sustainable...when the world maximum (with feedlot strategy having much much more room and per unit input than Bangladesh has) is around 30% (refer to US etc) and world average around half that and even lower in densely populated areas. This is why @bluesky stopped responding, the wall of logic hit too hard once you go all in with supporting BBS flim flam transaction counter = 1:1 individual unit stats (or whatever other method they are using to generate largest possible volume number).

There is a reason why end per capita surveying on consumption is best to gauge actual nutrition....raw supply side stats are notoriously bad when you do not have a good statistics organisation....and even if you do there are many other factors and efficiency coefficients in between that make it less useful than realised demand oriented survey.

 

[24 Hours]

Corruption perception Index.



Assuming North Korea to be the worst rock bottom. How far away is BD rank wise?



How so? They aren't even really measuring the same things.



The so called BBS statistics that claim massive increase in beef production from domestic herds. BBS claims that a 50%+ yearly herd slaughter rate is both realised and sustainable...when the world maximum (with feedlot strategy having much much more room and per unit input than Bangladesh has) is around 30% (refer to US etc) and world average around half that and even lower in densely populated areas. This is why @bluesky stopped responding, the wall of logic hit too hard once you go all in with supporting BBS flim flam transaction counter = 1:1 individual unit stats (or whatever other method they are using to generate largest possible volume number).

There is a reason why end per capita surveying on consumption is best to gauge actual nutrition....raw supply side stats are notoriously bad when you do not have a good statistics organisation....and even if you do there are many other factors and efficiency coefficients in between that make it less useful than realised demand oriented survey.

BD may seem close close to NK rank wise but theoretical score wise? No. NK is totalitarian to the max.

Corruption feels to complex to run down to a mere score. Truly we can expect difference among different parts of a single country to international borders. By looking at perceptions of corruption instead of actual corruption (which I'll cut them a break for makes sense given actual corruption is more difficult to asses). Cato's score deals with things that are more observable and can be applied to a more universal standard.

Anyway, this 50% stat, are you referring to what was posted in this?
Where it claims that the 50% of the cattle demand is consumed here.

 

[Nilgiri]

BD may seem close close to NK rank wise but theoretical score wise? No. NK is totalitarian to the max.

You are still in the vicinity compared to overall region (you pretty much bisect the space between India and North Korea in corruption perception index for example), thus your credibility is also by default affected. BD politics is a real sham, North Korea at least doesn't shy away from what it is...and owns it 100%. BD tries to project a persona that it simply has no factual grounding on (e.g election commission + court independence and crediblity)....in many ways that is worse.

Given the region itself overall has/had these tendencies, one can only imagine the relative scaling of this phenomenon w.r.t Bangladesh:

https://www.bloomberg.com/news/arti...ain-trips-help-india-study-impact-of-cash-ban

"Night light data captures informal economic activity, it is available at high levels of spatial disaggregation, it can be obtained in almost real time, it is relatively cheap to acquire, and it is not subject to politically-motivated interference."

Corruption feels to complex to run down to a mere score.

Hence the subcomponents in the CPI. We can discuss all of those and highlight any you feel are more relevant.

Truly we can expect difference among different parts of a single country to international borders.

So? That goes for any large scale measurement of almost anything. The average/median + assuming normal distribution from that tells us enough to compare the overall situation between countries. Not a real argument here.

By looking at perceptions of corruption instead of actual corruption (which I'll cut them a break for makes sense given actual corruption is more difficult to asses).

How do you even measure actual corruption? You would have to actually physically start destroying parts of an economy to get hard corruption/inefficiency data on it. Not a viable strategy.

CPI methodology/sources seem fine to me:

Methodology:

A) Reliable data collection and methodology from a credible institution: It is necessary that we trust the validity of the data we are using. As such, each source should originate from a professional institution that clearly documents its methods for data collection. These methods should be methodologically sound, for example, where an ‘expert opinion’ is being provided, we seek assurance on the qualifications of the expert or where a business survey is being conducted, that the survey sample is representative.

B) Data addresses corruption in the public sector: The question or analysis should relate to a perception of the level of corruption explicitly in the public sector. The question can relate to a defined ‘type’ of corruption (e.g. specifically petty corruption), and where appropriate, the effectiveness of corruption prevention as this can be used as a proxy for the perceived level of corruption in the country.

C) Quantitative granularity: The scales used by the data sources must allow for sufficient differentiation in the data (i.e., at least a four-point scale) on the perceived levels of corruption across countries so that it can be rescaled to the CPI’s 0-100 scale.

D) Cross country comparability: As the CPI ranks countries against each other, the source data must also be legitimately comparable between countries and not be country specific. The source should measure the same thing in each country scored, on the same scale.

E) Multi year data-set: We want to be able to compare a country’s score, and indeed the index in general, from one year to the next. Sources that capture corruption perceptions for a single point in time, but that are not designed to be repeated over time, are therefore excluded.

Sources:

13 data sources were used to construct the Corruption Perceptions Index 2016:

1. African Development Bank Governance Ratings 2015
2. Bertelsmann Foundation Sustainable Governance Indicators 2016
3. Bertelsmann Foundation Transformation Index 2016
4. Economist Intelligence Unit Country Risk Ratings 2016
5. Freedom House Nations in Transit 2016
6. Global Insight Country Risk Ratings 2015
7. IMD World Competitiveness Yearbook 2016
8. Political and Economic Risk Consultancy Asian Intelligence 2016
9. Political Risk Services International Country Risk Guide 2016
10. World Bank - Country Policy and Institutional Assessment 2015
11. World Economic Forum Executive Opinion Survey (EOS) 2016
12. World Justice Project Rule of Law Index 2016
13. Varieties of Democracy (VDEM) Project 2016

===================

Mind you tons of "actual" data w.r.t say poverty, development, health etc etc is also based on very much the same methodology of surveying/asking questions and relying on people's perceptions inherently (on say what they consumed in a timeframe) without direct physical auditing (best theoretically would also factor in a distribution of accuracy/honesty) etc. in the interest of generating the volume of data needed to hedge/give an appropriate C.I with low bias markers etc with the time/resources at hand.

If you have no problem with such data and its analysis, then there is little to inherently have with the CPI index either which really aggregates and standardises markers from 13 spread out sources (which in turn take from other standardised base surveys and/or reports...adding much hedging and reduction of bias in the process).

Anyway, this 50% stat, are you referring to what was posted in this?
Where it claims that the 50% of the cattle demand is consumed here.

Reading it again, something even more unbelievable is implied:

During the sacrifice feast of Eid-ul-Azha, Bangladesh consumes 50% of the country’s annual demand by sacrificing around 8.8 million cattle .

If 8.8 million is 50% of the annual demand (generated just by Eid). That implies around 18 million head of cattle is the claimed total annual demand.

But then BD reports 24 million or so total herd size and now only an insignificant amount (0.5 mil from india, 0.X mil from further away - cant be bothered to look it up) imported to top it off:

http://www.dhakatribune.com/banglad...rices-sacrificial-cattle-may-soar-due-floods/

A DLS source said there were 23.8 million cows and buffalo

18/24 = 75% claimed processing rate.

World max hovers around 30% in comparison, World average around 15% and high pop density average much less (often in single digits) - if you want numbers I can provide them (already done in the case of the first one w.r.t USA).

Please tell me the phenomenally awesome animal husbandry rates and <12 month growth rates achieved only in Bangladesh compared to the world for a sustainable herd size in such an extremely densely populated + feedlot restricted area + low unit input to boot. Hopefully its scaleable/applicable to all other animal rearing/production/processing so we can increase world production by 6 - 7 times by simply following the awesome, revolutionary BD discovery.

Or is Occam's razor that something fundamentally wrong with the BBS claims more applicable? You decide.

I mean people are already using these numbers as established fact of some crazy y.o.y growth never done in history anywhere else in the world:

https://defence.pk/pdf/threads/bang...-consumption-soars.525749/page-2#post-9979736

@madokafc @Aung Zaya

 

[BitHeroBD]

BD is worse than NK

 

[Nilgiri]

140 Million out of 160 Million, when 55 Million of the population is under the age of 15 is impossible. Either your country is under-reporting your population or you are counting multiple subscriptions by individuals as separate users.

The only other possibility is that every 10 year old and above in your country owns a mobile. With 43 percent of your population living under the poverty line, it does not seem possible that every household can afford to provide their 10 year olds with a cell phone.

 

[Nilgiri]

http://www.thedailystar.net/frontpage/record-growth-tough-year-1492489

Yesterday, WB economist Zahid said, “The larger puzzle -- the inconsistency between the 7.3 percent GDP growth estimate and the growth related high frequency indicators -- remains unresolved.”

He further said, “Public investment to the GDP has increased by 0.75 percentage points in a year when the ADP implementation rate is 3 percentage points lower than last year.

“How could manufacturing growth be nearly 11 percent in a year when LC settlement for import of industrial raw materials grew by only 3.5 percent and exports in nominal dollar terms increased by only 1.7 percent? How could construction grow by nearly 8.8 percent when LC settlement for the import of construction materials declined by 0.5 percent,” he asked.

“There are also directional inconsistencies. For instance, growth in wholesale and retail trade, transport and real estate sectors increased relative to last year while growth of credit to the private sector decreased.”

The WB economist pointed out, “This does not necessarily mean that the reported numbers are wrong. But they do raise a lot of questions which the final estimates do not help answer.”

=====

Thank you to @HAKIKAT

 

[Nilgiri]

Sometimes. international statistics people cannot understand the reality in BD. For example, consumption of animal meat during two or other occasions. If the 80% of the hides are smuggled out to India how can the international people will count the real meat consumption in BD?

Hey idiot, do you understand how long it takes Zebu cattle (and all cattle in general) to grow from birth to full size (for beef production, forget dairy for now)? Hint: its not under one year. Combine that with nowhere near 100% animal husbandry rate (that adds another coefficient). Factor in feedlot and input costs and efficiencies after that. It has been proven that you can reach only 30% for beef production/herd in several countries (that dont have a stupid BBS blah blahing) at best...that too with huge amount of room and input/animal (that BD has next to zero). If anything the rate in BD is probably around 10%, yet it claims 75% (probably because it measures some form of lax market transactions rather than actual individual units...and hope no one notices that is never 1:1 ratio.... for biggest possible feel good number for domestic morons so they vote for BAL because "improvement").

When you slaughter 75% of a herd (of pretty much any large animal) you have in just a year, you will have 0 herd incredibly fast. DO YOU UNDERSTAND THAT BASIC THING?

Sorry BBS thinking it can lift chicken rearing rates or something of that size to cattle rearing rates as though they are the same growth and feed input time cycles is the most stupidest thing ever, even for a propaganda outlet.

If you personally lack the basic aptitude to figure out this basic logic that BBS numbers are basically what the last two letters in acronym are....then I have no idea what you are doing (supposedly) in Japan....past being an illegal dhobi kind of fellow there like you BD's are sooooo good at.

"International statistics" and "consumption of meat during two or other occasions" (BBS already claimed eid is 50% of the claimed total demand) ....has ZERO to do with the BBS internally derived "results" being a bunch of bullcrap when basic cattle rearing fundamental rates (backed by a plethora of non BBS world data) are applied.

Super rich, super big country with massive beef production (say USA), has 30% sustainable yearly processing rate of herd....backed up by rates of every other country in the world (which are all about same or lower).

Poor impoverished, low investment and super densely populated, low area BD has 75% processing rate because super credible BBS says so....yeah ok

 

[bluesky]

@Nilgiri, do not you really believe that BD economy grew by 7.2% last fiscal although ADB, IMF, WB etc. accept it as the real figure?

 

[Nilgiri]

@Nilgiri, do not you really believe that BD economy grew by 7.2% last fiscal although ADB, IMF, WB etc. accept it as the real figure?

It will be around that general vicinity sure (imprint on real nominal growth in USD). BD is growing its PPP for example quite ok....and overall GCF is stabilised at decent level. BD does subscribe to GDDS format in IMF, it has certain standards to follow and open to vetting on the major ones (liquidity flows esp with BoP which is easily cross checked by other major partner countries) so BBS runs a more tight ship there (it has to) so not a total devoid of logic zero by any stretch there.

At this stage of economic growth too, there will be discrepancies among a bunch of high frequency markers (with official growth figure and subcomponents), but precisely how that trend is going w.r.t recent past in BD, I have not looked into (because it should generally improve or stay about same than decline if your liquidity base/exposure/generation improves...which one World Bank economist was implying as a troubling issue cropping up now....because decline suggests informal is expanding bigger than formal economy and thats not good in long run).

The bigger question is anyway how much that (growth figure whether 4,5,6 or 7 percent etc) has an impact on genuine underlying socio-economic improvement in BD, given BBS has much bigger buffer there to be exploited for political reasons. Those are where the larger discrepancies are found because they are not high frequency, neither subscribe to same level of vetting/standards process under IMF and others.

 

[Nilgiri]

https://www.thedailystar.net/business/banking/wb-doubts-765pc-gdp-growth-estimate-1560541

With thanks to @Ashes :

https://opinion.bdnews24.com/2017/12/18/where-did-the-benefits-of-economic-growth-disappear/

There is some disquiet among economists about the quality of data provided by Bangladesh Bureau of Statistics (BBS). No less a person than the Economic Adviser to the Prime Minister has expressed his dissatisfaction about the qualifications of the BBS staff and the quality of their work. These are very long standing problems, and yet, very little has been done to improve the quality of the services provided by BBS.

I will analyse over time this article @Joe Shearer which highlights some of the issues I had talked about earlier in various threads with regards to bad state of BBS.

 

[Nilgiri]

https://defence.pk/pdf/threads/wb-doubts-7-65pc-gdp-growth-estimate.552983/#post-10403712

https://defence.pk/pdf/threads/bang...apita-1-733-usd-in-2018.554189/#post-10422781

 

[Nilgiri]

This could potentially be very big for BD and region...esp in context of what I have been saying in this thread and larger subforum:

https://www.channelnewsasia.com/new...-corruption-policy-for-member-states-10165016

IMF unveils new corruption policy for member states

The stricter new policy also aims to tackle how rich countries contribute to corruption in the developing world by failing to prevent bribery and money laundering or by allowing anonymous corporate ownership.

WASHINGTON: The International Monetary Fund will systematically address corruption and its impact on economic growth with all its member countries under new guidelines launched on Sunday (Apr 22).

The stricter new policy also aims to tackle how rich countries contribute to corruption in the developing world by failing to prevent bribery and money laundering or by allowing anonymous corporate ownership.

"We know that corruption hurts the poor, hinders economic opportunity and social mobility, undermines trust in institutions and causes social cohesion to unravel," IMF Managing Director Christine Lagarde said in a statement.

"We have now adopted a framework for enhanced engagement on governance and corruption that aims for a more systematic, evenhanded, effective and candid engagement with member countries."

Corruption and poor governance sap economic growth and exacerbate inequality, according to the IMF, and the new policy framework ensures the institutions will hold all members to the same standards - something it had not always done.

The new policy comes as Ukrainian authorities work to implement stringent new anti-corruption reforms at the behest of the IMF, which has held up the latest instalment of a US$17.5 billion aid package.


The revised good governance guidelines, which take effect on Jul 1, follows a recent review of the IMF's 20-year-old policy framework which concluded the fund had sometimes employed euphemisms when discussing corruption in member states - leaving local officials unclear about IMF concerns.

'MORE INTRUSIVE'

And IMF analysis sometimes failed to apply the same standards evenly to all members.

Under the new guidelines approved by the IMF board on Apr 6, the fund will discuss good governance concerns in all annual economic reviews of member countries.

IMF officials however say they do not expect the policy will lead to more stringent conditions on loans, which go to a minority of the fund's 189 members and which already include anti-corruption provisions.

The fund also will rely on the findings of outside transparency campaigners who have criticised the existence of tax and corporate havens in advanced economies as a conduit for illicit financial flows to and from poorer countries.

However, the IMF will not investigate specific instances of corruption.

Rather it will focus on the strength of key economic institutions: fiscal and central bank governance, market regulation, the rule of law and policies on money laundering and countering terrorism financing.

IMF analysis suggests falling 25 notches on a corruption index could shave as much as 0.5 percentage points off a country's annual growth - amounting to tremendous economic losses over multiple years.

In unveiling the new policy on Sunday at the close of the IMF-World Bank spring meetings, Lagarde said the fund's board supported a "more intrusive" approach to member state evaluations.

"Because it is a macroeconomic issue, the IMF is really perfectly legitimate in acting, especially when we have a program in a country," she said.

Source: AFP/de
Read more at https://www.channelnewsasia.com/new...-corruption-policy-for-member-states-10165016

@Joe Shearer @bluesky @Gibbs @Ashes @Mage

 

[Nilgiri]

@Chak Bamu @Major Sam @A.A. Khan @Areesh @Cookie Monster

Continuation of: https://defence.pk/pdf/threads/bang...11bn-at-end-april.557485/page-3#post-10479510

I'm not sure what to make out of this calorie intake.

It is from BBS, just like all the other stuff being trumpeted when narrative is convenient. Why pick and choose which feels right and wrong?

.It could be 2010 data was a fraud?....I wouldn't know...It's not like they started data fudging just now.

Actually the sample size for 2016 HIES is 4 times larger than the 2010 one. It could be the 2016 one simply captures a more wholesome picture.

HIES 2016:

http://documents.worldbank.org/curated/en/480321488461618499/pdf/WPS7989.pdf

Sampling size around 50,000 (scaled up 4 times from 2010).

Better weights, lower bias too for 2016? It looks that way. But either way the decline is very worrying if you are picking up such by increasing sample size and increasing accuracy.

Bangladesh still does better than Pakistan and INdia in hunger index.

Largely because Bangladesh is decreasing its sample sizes for such (malnutirtion, hunger, health and mortality) whereas India and Pakistan are increasing them (and improving):

https://data.unicef.org/resources/j...estimates-2017-edition-interactive-dashboard/

Survey size collapse for BD lately...why are India and Pakistan overall increasing their survey sizes (this factors into improving in corrpution +institutional perception indices btw) but BD has halted helen keller mass survey size of the 90s etc when it comes to malnutrition etc?

What you think the result will be if BD does a proper scale survey like used to be done, and on the scale of HIES (50,000 people surveyed instead of a few thousand?). Shouldn't BD be increasing its sample sizes (like India and Pakistan are overall) instead of decreasing them especially when it has a corruption perception/institution credibility problem?

Also remember sensitivity/weightage (compared to non poor) of the poor/marginal (esp children) to decline in calorie intakes w.r.t stunting etc (which only will show up in later years)...i.e they are much much more vulnerable given they are much closer (or already below) to the threshold lines of intense poverty.

Not to mention looking at calorie intake by country.....it says Mauritania's per capita calorie intake is higher than that of Japan....Niger scores higher than India.

Mauritania and Niger have statistically sound govt institutions for this measuring? You are exactly proving my point when it comes to Bangladesh....the more I dig into BD sample surveying process, the more disturbing things get from credibility standpoint (like the aforementioned sample size reductions). I am not going to be surprised if I dig and find this same trend when it comes to education and other socio-economic surveying of BD.

Reduce/control the incoming data flow because it helps the end stats generally.....whatever the hit on credibility...because much fewer people are interested in analysing the latter (strategy) compared to former (tactics).

Not to mention based on 2008 data, Pakistan's calorie intake was just higher by 10 calorie despite them having more than twice PPP income than Bangladesh back then.....They say calorie intake decreased because of decrease in rice intake....

Again Pakistan seems to have more credible surveying trends than BD. When you look at larger overall analysis regarding food security:

2012:

BD = rank 81/105 (23% percentile) (Ind = 37% Pak = 29%)

2017:

BD = rank 89/113 (21% percentile) (Ind = 35%, Pak = 32%)

http://foodsecurityindex.eiu.com/Country/Details#Bangladesh

http://foodsecurityindex.eiu.com/Ho...urity Index - 2012 Findings & Methodology.pdf

Food security index is also notably higher multi-variance input, lower sensitivity to survey bias and smaller sample sizes etc compared to GHI. It's not surprising that BD is doing worse on it from an already bad position.
Correlates with the worsening calorie intake position BBS itself has stumbled upon.

Bangladesh is one of only handful of countries that "needs improvement" in Asia:

Given BD performance when much more multi-variate analysis (and broader scope of survey collections) is employed, it is disingenuous to use only GHI (esp when higher weight higher sample size studies like what would show up if HIES 2016 is more correct in its capture.... are not available, because BD seems to find that uncomfortable).

As for it being "less rice intake" based, well that is a poor argument when the calorie intake average is so low to begin with?

It's hard to believe.

Well BBS surveys magnitude more than you do.

salary of govt employees increased a lot in last 6-7 years.

and? I'm sure the rich in BD got richer too. But they are 0.1% or less of the population.

I have said before,based on my observation....things have improved.......i.e. begging, rickshaw fare....Obviously more rickshaw puller now owns a mobile phone compared to 2010. Obviously more people using Internet compared to 2010.

Again you didnt survey 50,000 people in random fashion like BBS supposedly did.

More cars running in the streets compared to 2010....these things cost. If income were going down I wouldn't expect it to be that way.

Again you seem to be ignoring that incomes could be rising just for the elite that can afford such things as cars and personal transport in the first place. In fact that is exactly what the BBS HIES 2016 suggests.

The question is the correlation/relevance of every consumption marker to the actual meat of the socio-economic pyramid.