This article argues that the understanding of the Bangladesh economy is distorted by the failure to include the full extent of the flow of remittances into the economy.
While the formal economy of manufacturing and related service faces many problems, the rural economy, we argue, is in much better condition.
Data is difficult to obtain and one must patch together pieces of information. The first part of the article estimates the flow of remittances. The second part discusses the use of the foreign exchange in the hundi market and the implications for understanding the economy.
How large is the inflow of remittances?
What is the total value of remittances that come to Bangladesh? I have asked many economists this question and the answers range from $30 billion to $60bn.
The amount through the official channel is about $25bn so the implied informal (hundi) amount ranges from $5bn to $35bn.
To make an estimate we try to determine how many persons are sending remittances and the annual amount that a person sends.
The estimates of the number of persons who are working abroad is about 10-15 million. We use 15 million. We use the high end as it is very difficult to track workers coming and going and there are many illegal workers not counted.
BBS in the 2022 Household Income and Expenditure Survey estimated 8.3% of households receive remittances from abroad. BBS reports 38.3 million households for the country.
This implies 3 million families receive remittances. Allowing for multiple workers per household we might reach 3.2 million persons sending remittances from the 2022 HIES estimate. This is far too low.
The BBS survey provides an estimate of an average remittance of Tk257,000 or $3000 per annum [ER 86 Taka/dollar (2022)]. An estimate of the average remittance from a bank over the past year (2024) resulted in an estimate of $4,000.
Below we will estimate the 2022 balance of payments so we will use the BBS estimate. With the 15 million workers abroad we obtain in 2022 a total remittance volume of $45bn, $22bn through the banking sector and $23bn through the hundi market.
Note if we use the HIES estimate for the number of recipients then the total remittances come to about $9.6bn ($3X3.2bn) which is much less than what passes through the formal financial sector.
In 2024 we use $4,000 per remitter (the bank's estimates are in dollars so we do not have to worry about the exchange rate) and increase workers to 15.5 million.
This suggests total remittance of $62bn; the hundi total is $38bn and the amount the financial sector we estimate as $24bn based on nine months data for 2024.
This indicates a large increase through the hundi system in 2024. This is due to: (a) Increase flow of overseas workers and (b) greater remittances for overseas workers due to inflation in Bangladesh.
The conclusion is that the economy receives a very large amount of funding through remittances, with more than half coming through the hundi market.
For 2024 we forecast this remittance amount ($62bn) is about the same as we forecast of the exports of goods ($60bn). The forecast for 2024 may be high as the $4,000 is an increase of $1,000 over 2022.
Impact of the hundi system
The current level of inflow of remittances $60bn per annum provides substantial funding for the low income part of the population and eases the impact of poverty. Inflation’s impact has been reduced by the total remittance flows and the unequal income distribution of households mitigated.
Currently there is considerable economic activity in rural areas, with construction of housing, purchase of farm equipment, and improvement of fields. Labour is in short supply in many rural areas. These are a consequence of the $50bn of remittances we estimate goes to rural areas, the remaining $12bn to urban areas.
The balance of payments looks rather different with a much larger flow of imports and a much larger inflow of remittances. In addition, there is a small outflow of capital.
We adjust the 2022 balance of payments accordingly.
The FY22 balance of payments might look as follows:
Adjusted balance of payments FY22: (bn$)
Notes: The official balance of payments is adjusted as follows:
- a) Imports are increased $20bn to include under-invoicing. Our procedure is as follows:
Under-invoicing of the imports we take at 20%. Our estimates of under-invoicing comes from the PSI work. Total under-invoicing in 2022 we estimate at 20%. 20% of imports of $96.1bn is about $20bn.
Estimating under-invoicing in 2022
[% of Imports, Export Promotion Bureau; % of Under-invoicing from PSI Data Contribution to Under-invoicing column 1 and column 2];
- b) Remittances [secondary income] are adjusted plus $23bn for the hundi inflow and minus $2bn for outward remittances to other South Asian countries;
- c) Capital flight through the hundi market is then approximately $1bn.
This version of the balance of payments is based on the hundi market providing $26bn in addition to the remittances through the formal financial institutions of $21.7bn. This total of $47.7bn is the amount that is in the hands of the Bangladeshi recipients.
What is the use of the foreign exchange corresponding to this $47.7bn received by the households? $21.7bn came through the financial system and is used for the various purposes of the economy (mostly imports). The demand for the hundi market foreign exchange component is divided into three categories: Outgoing remittances, under-invoicing of imports, and capital flight.
The trade documentation in Bangladesh has many problems. The primary problem is under-invoicing of imports. The several years of Pre-Shipment Inspection [PSI] operations revealed how pervasive this phenomena was. Imports from China and India in particular were found to have large widespread under-invoicing of imports. The high import tariffs (including VAT) motivates the under-invoicing.
For China the PSI reported every invoice was under invoicing by 100%. This meant actual imports were twice official. However, we estimated PSI coverage was 60% of imports.
India PSI estimates of under invoicing were 25% of official. Although the PSI operations have stopped one would expect as a result more under-invoicing not less.
Under-invoicing: The invoice for the import was split into two parts. One part would be paid through an L/C and the second part would be paid through the hundi system. In one memorable case the importer challenged the PSI Company’s finding that the actual price was higher than reported on the L/C. When the case was heard the PSI Company produced a document given to them by the exporter that listed payment as part L/C and part bank transfer.
There are several consequences of pervasive under-invoicing:
- a) The government loses a lot of revenue.
- b) Importers may choose a source of imports that is more expensive in dollar terms hence costing the nation more foreign exchange for the import than it needs to pay.
- c) It leads to corruption of the customs service. The customs department hated the PSI and worked hard to discredit it and stop the program. They succeeded. The importers also hated the PSI system and preferred to pay a customs officer than the official amount of duty and VAT they would have to pay as the under-invoicing was exposed by the PSI system.
There are cases of over-invoicing of imports when the duty is low so as to achieve capital flight. The PSI program was trying to establish the legal amounts of duty and VAT to pay on imports. Customs was not interested in preventing capital flight. This over-invoicing was common for capital goods leading to an overestimate of investment in the national accounts. The rental power plants were a notorious example as these are hard to value and were often overvalued; this resulted in high cost electricity.
Seen in this way, the Bangladesh economy is certainly doing better than most think.
First of all, there is a tremendous amount of remittance funds in the hands of the recipients. Of course it is concentrated in about 10% of the rural households. But these households are spending this money for construction and purchasing goods that are largely produced in the region. Hence, the remittances have a multiplier impact so others in the area benefit from the expenditures of the recipients of the remittances.
Informal discussions suggest that the current conditions in rural Bangladesh are not bad: There is limited unemployment, people have the money to get by although high food inflation makes things tougher than usual. In 2024 the remittance amount increased to around $62bn giving a stronger impulse to the growth and economic welfare in the rural areas.
When the official data indicate a decline of imports this may be a false signal as importers may increase the amount of financing through under invoicing. This would result in less capital flight. But it will also cause the exchange rate to depreciate as the demand for under-invoicing dollar funds drives down the value of the Taka.
Forrest Cookson is an economist who has served as the first president of AmCham and has been a consultant for the Bangladesh Bureau of Statistics.
