Syndicate content

Why taxing remittances is a bad idea

Dilip Ratha's picture
In 2016, migrant remittance flows to developing countries amounted to $440 billion, more than three times the size of official development aid flows. In many countries, remittances are the largest source of foreign exchange. In India and Mexico, they are larger than foreign direct investment; in Egypt, they are larger than the revenue from Suez Canal; and in Pakistan, they are larger than the country’s international reserves.

Recently, a number of rich countries that host a large number of migrants are considering taxation of outward remittances, in part to raise revenue, and in part to discourage undocumented migrants. The list of countries where such taxes are being considered includes Bahrain, Kuwait, Oman, Saudi Arabia, the United States, and the United Arab Emirates. (In the United States, Oklahoma taxes remittances at the rate of $5 for the first $500 and 1% thereafter. Two other states, Georgia and Iowa, are considering taxes that may have a wider scope by taxing not just remittances but also other transfers. See Mohr (2016)).

We outline below nine reasons why taxing outward remittance flows is a bad idea:
  1. Since the income of migrants has, in principle, already been taxed in the host country, taxing remittances amounts to double taxation for tax-paying migrants. Since remittances are usually sent to poor families of migrants, the tax would be born ultimately by them and therefore it is likely to be highly regressive.
  2. A tax on remittances will raise the cost of remittances, in direct contravention to the G20 commitments and the Sustainable Development Goal of reducing remittance costs and increasing financial inclusion.
  3. Poor migrant workers tend to be highly sensitive to the costs of remittances. A tax on remittances will drive these flows to unregulated, informal channels. That is likely to reduce the tax revenue, increase the cost of tax administration, and encourage informal channels of money flows, raising security risks.
  4. To the extent that remittance channels are used also for small-value transfers for purposes of trade, tourism, investment and philanthropy, these latter variables will also be impacted by a tax on remittances.
  5. A tax on remittances, especially if it is applied selectively to the nationals of a country, can redirect flows through third countries. (Anecdotally, a U.S. ban on remittances to Iran has forced Iranians in the United States to send money through Europe or the UAE.) If so, migrants will have to pay remittance fees twice.  
  6. Estimates suggest that the revenue raised from a tax on remittances will be small relative to the revenue base of the country. For example, IMF estimates that a remittance tax of 5 percent would result in revenue of around $4 billion or 0.3 percent of GDP of the GCC countries (IMF 2016). United States Government Accountability Office simulations suggest that a potential fine of 7 percent on remitters without legal status in the United States would raise less than $1 billion in revenue, and chances are the revenue would fall below the cost of tax administration to administer and enforce the tax.
  7. A tax on remittances would affect the volume of business of remittance service providers, thereby reducing their tax payments. 
  8. A tax on remittances may contribute to driving expatriate employees and entrepreneurs to other countries with lower taxes.
  9. In the past, such taxes have not worked. In Gabon (in 2008) and Palau (in 2013), tax collections were found to be insignificant (IMF 2016).
In the past, many developing countries have been tempted to tax inward flows of remittances, but in the end, very few countries actually did. The drawbacks of taxing inward remittances are similar to those of taxing outward flows. Taxes can drive remittances to informal channels, making tax collection difficult and costly (Mohapatra and others 2012). They also impact poor families disproportionately.
Indeed, a few countries that had such taxes on inward remittances ended up removing them. Vietnam removed its 5 percent tax on remittances in 1997 and found that remittances through formal channels increased. Removal of Tajikistan’s state tax on cross-border bank transactions in 2003 may have helped raise formal remittances from $78 million in 2002 to $256 million in 2003. In the Philippines remittances were subjected to a document stamp tax. Since 1995 this is exempt for transfers by overseas Filipino workers on presentation of appropriate documentation. India imposes a small service tax on the fees charged by money transfer agents, but not on the remittance amount.
There is a need for a systematic study of feasibility and implications of taxation of (outward and inward) remittance flows. Such a study would involve preparation of country case studies including interviews with remittance service providers, migrants, their households back home and tax authorities. Since the literature on taxation of remittances is largely silent, there is a need for analytical modeling of such taxes, perhaps drawing on the literature on taxation of charitable contributions. 
Business Mirror. 2015. “OFW remittances and the exemption from documentary stamp tax.” (Link)

Business Today. 2014. “Sending money back home to India gets costlier for NRIs” (Link)
Cuevas-Mohr, Hugo. 2016. “The Taxing of Remittances in the US.” IMTC. (Link)
IMF. 2016. “Diversifying Government Revenue in the GCC: Next Steps.” Note for Annual Meeting of Minister of Finance and Central Bank Governors of GCC. (Link)
Mohapatra, Sanket, Blanca Moreno-Dodson, and Dilip Ratha. 2012. “Migration, Taxation, and Inequality.” Economic Premise, May 2012, Number 80 (PREM), World Bank. (Link)
United States Government Accountability Office. 2016. “International Remittances: Actions Needed to Address Unreliable Official U.S. Estimate.” February. (Link)


Submitted by Abhishek Saurav on

Informative note Dilip, I enjoyed reading it!
As micro-transfers international remittances have transformative potential. In burdening remittances through taxation, governments impede rather than leverage their potential. In fact, policy should recognize the productive possibilities of stable and timely household monetary flows. It can magnify its use in spurring growth in lagging regions, fostering spatially balanced job-creation, and unlocking latent entrepreneurship capacity in family members who are left behind.

Submitted by Kaushik Ghosh on

Dear Mr. Ratha,

I follow your reports on remittances furnished on the WB website and for anyone working in this industry, it has always remained our Bible. Was curious to understand one thing though, when you mention that remittances to India in 2015 was $71 bn, are all of it family remittances? because today there are not even formal channels to send money from countries like Bangladesh, Pakistan and Sri Lanka to India which constitutes a major chunk of the $71 bn.

Look forward to you hearing from you. You may reach out to me on my email or IM me on my LinkedIn (we are connected).


Thanks for your comment, Kaushik. Typically the officially recorded data on remittances tend to underestimate the true size, among other reasons, due to the difficulties of measuring flows through informal channels. When the cost of sending money Is high, or difficult due to exchange controls imposed by the governments on either side, migrants are forced to use informal channels. A discussion of the definitions, methods and pitfalls of measuring remittances can be found on page viii of the Migration and Remittances Factbook 2016. Kindly see link:

Submitted by Charles Farley on

U.S. Representative Mike Rogers apparently did not read your blog, Dilip. On March 30 he introduced a bill to impose a 2% tax on remittances sent from the U.S. to any country in Latin America and the Caribbean. Revenue will be used to build The Wall along the southern U.S. border.

the text of the bill is at:

In a news piece, his staff pointed to Haiti and other LAC countries where remittances are large (>15%) of GDP as ones that benefit from porous U.S. borders and should therefore be made to pay for enhanced border protection. (See link )

Add new comment