Ever since the people of the country started migrating to other countries in search of employment and better livelihood, they remitted money to their families at home. These remittances soon became an important prop to our balance of payments (BoP) frequently amounting to much more than one-half of the stock of international reserves.
It is alleged that a large amount of remittances is brought into the country by hundi transactions. One of the most firmly entrenched conventional wisdoms in popular economic discourse in Bangladesh is that hundi is a 'bad thing'. Like all such wisdoms it is based on apparently plausible arguments that are easy to understand and communicate. Hence, it has wide acceptance even among the policy makers. However, conventional wisdoms have been found to be not always correct. This write-up investigates if the oft-repeated allegation about hundi is correctly constructed.
In modern times, hundi refers to an informal financial arrangement to transfer money from one country to another. It usually involves a conversion of the currency of the sending country to the currency of the recipient country. It arises when there are significant distortions in the foreign exchange market such as currency overvaluation. It bypasses the formal financial system entirely such that there is no record of such transactions such that the government agencies cannot trace them. Hence, there is no record of the total amount of money brought in or taken out of the country through hundi. However, it is widely believed that a substantial part of the worker remittances is brought in using hundi facilities.
A major attraction of hundi is that it is a very simple and convenient way to transfer money from overseas in comparison to the formal alternatives. Suppose an expatriate worker in Dubai wants to send US$1,000 (or its equivalent in Dirham) to his mother at home. He contacts an agent of a hundi merchant (assume he is not a Bangladeshi). He gives the agent the necessary information. They also agree on the exchange rate, say Tk120 per US dollar. Shortly thereafter, the agent meets the worker again and calls the mother on the phone. He asks the worker to confirm on the phone if his mother has received Tk120,000. When the mother confirms the receipt, the worker pays US$1,000 to the agent, and the deal is closed. No paper work is required which is very convenient if any of them is not sufficiently literate. The transaction is confidential, instantly verified and hassle-free. An additional benefit is that the taka equivalent of the dollar in hundi transactions is usually higher than what the formal institutions would give for a similar transfer.
The preceding discussion suggests that the remitters of money and their beneficiaries at home are unlikely to regard hundi as a bad thing since they have voluntarily chosen it over the formal channels. But there is a relatively small group of influential people who have constructed and persisted with the notion that hundi is a bad thing. Since the media propagates mostly their opinion, it has entrenched itself as the conventional wisdom in this regard.
A systematic economic analysis of why hundi might be a bad thing is very hard to come by. However, the following arguments are often made in public discourse: (a) Hundi is not a legal instrument for money transfer, and hence its use is unconscionable; (b) there is no written record of hundi transactions which makes detection and policy planning difficult; (c) if any foreign exchange is remitted through hundi facilities, the beneficiaries receive the taka equivalent of this amount. They never receive any foreign exchange. In fact, the foreign exchange may never reach the shores of Bangladesh. Hence, the country is deprived of the benefits of a substantial amount of foreign exchange earnings of our workers. If the money were remitted formally through banks, it would have raised the international reserves and import capacity of the country substantially. This would have improved the balance of payments, increased the international reserves, and strengthened the taka exchange rate.
The first two of these arguments are held mostly by the ardent admirers and beneficiaries of dirigisme. These are essentially about government control which may not have much to do with promoting economic welfare of the workers. These will not be discussed here. But the third argument apparently entails important economic implications and hence, warrants scrutiny.
There is no question that neither the beneficiary of the remitter at home nor any local bank receives the remittance in foreign exchange in a hundi transaction. This fact has been interpreted as a loss of the foreign exchange to the domestic economy. But does it really entail a loss of the benefits of the foreign exchange earnings of the worker?
A shortcoming of this 'loss' argument is that it does not ask how the hundi merchant in Dubai acquired Tk120,000 to pay the mother of the worker. The only way a foreign entity can have access to funds in Bangladeshi currency is through the export of some goods, services or assets to Bangladesh. Hence, the hundi merchant must have either exported something to Bangladesh directly or, what is more likely, did so indirectly through the foreign exchange market. This implies Bangladesh had utilised or had the opportunity to utilise the foreign exchange earnings of the expatriate worker even before the money was received by the mother. Hence, if the worker's earnings were sent instead through the formal channels it would not have materially affected either the import or the reserves of the country, and consequently would not have much impact on the exchange rate.
If the hundi merchant obtains taka by selling an asset, such as a deposit in a bank account in Dubai, then the foreign exchange need not be sent to Bangladesh. In this case the transaction could be in the nature of a capital flight.
If the hundi merchant is a resident Bangladeshi businessman with agents in Dubai, the money transmission process is the same, but since the hundi merchant may also have other sources of revenue in Bangladesh, he may not engage in any counter transactions to convert the remittance money to taka to pay the mother. He could do it from his local earnings, and instruct the agent in Dubai to deposit the foreign exchange in his bank account there. In this case neither the foreign exchange nor any import comes to Bangladesh. If this transaction were done formally, it would be shown as an increase in the foreign assets of a Bangladeshi business, which is offset by an equivalent reduction in assets at home. But if the transaction is done off the record, as it is in the case of hundi, it is regarded as illicit capital outflow. In the event the merchant does not have sufficient funds locally, he would have to pay the taka amount by remitting the foreign exchange home. In this case the loss of foreign exchange argument is not relevant.
The correctness of the conventional wisdom regarding hundi depends on whether or not most of the hundi transactions are done by Bangladeshi merchants. It is believed by many that the majority of the hundi transactions is done by non-Bangladeshi merchants. If this is so, hundi should not entail a large foreign exchange loss to the domestic economy.
An important issue here is that when the remittances are sent through the formal channels, the government may determine the usage of the foreign exchange as well as the exchange rate. Importantly, it can decide on what can or cannot be imported. But when the fund is transferred through hundi, it is the informal market that determines how the foreign exchange would be utilised and at what rate. The government loses control over the disposition of a part of the foreign exchange earnings of the nation and a parallel exchange market develops that determines a kerb market rate of the foreign exchange.
The discussion above suggests that the effect of remittances through hundi on the domestic foreign exchange market is much more nuanced than what the conventional wisdom would have us believe. Consequently, even when the government succeeds in fully eradicating hundi, it may not have much impact on the international reserves or the exchange rate and thus the effort may not be very rewarding. However, it will have some significant distributional consequences such as the higher premium now received by the beneficiaries of the remittances will accrue to the central and commercial banks, importers and perhaps government functionaries.
The discussion above focuses on (in-bound) hundi used by our workers overseas. It does not explore the important case of (out-bound) hundi whereby locally earned taka is converted into a foreign currency and sent overseas through informal channels, or money outflow through misinvoicing. These are also illicit financial outflows. Global Financial Integrity (GFI) reports suggest that much more money is transacted through this type of hundi than the one discussed above.
It should be borne in mind that hundi exists only because the government represses the exchange market and trade. The taka has been almost always considerably overvalued and the domestic industries have been ring-fenced by high tariff walls. These measures were taken ostensibly to reduce the current account deficit and also promote industrialisation. However, these actually discouraged export industries and encouraged imports causing an increasingly large current account deficit, which ultimately led to a greater overvaluation of the taka and greater repression of the exchange market. This in turn made hundi even more lucrative. If the government is really serious about reducing the use of hundi, it should make the exchange rates more flexible, and trade and capital flows more liberal.
The author is a Professor of Economics, Independent University, Bangladesh. m_a_taslim@yahoo.com