This blog post is joint work with Jeetendra.
One important part of capital account decontrol is commitment. If there is risk that capital controls will be brought back in the future, this can have a variety of unpleasant effects. If there is a fear of fresh restrictions coming in on inflows, a surge of money will rush into the country. If there is a fear of fresh restrictions coming in on outflows, a surge of money will rush out of the country. A long-term commitment to openness is required, in order to rule out such behaviour.
As a consequence, when a country moves to full convertibility, this requires not just the removal of restrictions. It also requires the removal of bureacratic process including reporting requirements. As long as forms have to be filled up for `automatic approval', this can easily swing back and become a capital control through breakdowns of rule of law (as has happened in India). See the MoF Working Group on Foreign Investment on issues of rule of law in India's capital controls. It is important to pour concrete on the decontrol so as to give confidence that the controls are gone.
We don't have the exact facts, but in the UK, when they moved to convertibility (back in the late 1970s) this was accompanied by dismantling of reporting requirements.
Korea is very open; there are no restrictions on capital flows. But Korea has kept the reporting requirements and through this, they have retained controls in a certain sense. The reason is that people fear that if they report transactions, then the government may come and investigate, and ask why they are doing it. They might also ask where the money is coming from. So, even though the rules may allow capital transactions, people -- especially individuals, but also small businesses -- remain very wary of these, and refrain from wiring large amounts in and out of the country, apart from some outward investments via mutual funds, where the government can't actually see who is sending the money out. Through this, reporting requirements perpetuate home bias and inhibit international economic integration.
Today we became aware of one mechanism through which some countries have committed themselves to an open economic system: When the US signs free trade agreements and bilateral investment treaties, there are provisions which limit the extent to which capital controls can then be brought back.
A curious letter has brought this to our notice. It says: Under these agreements, private foreign investors have the power to effectively sue governments in international tribunals over alleged violations of these provisions.. How interesting. So that locks down the possibility of a reversal of reforms in countries where the US has free trade agreements, and quite a few more where the US has bilateral investment treaties.
It makes sense for investment and trade treaties to mention capital controls. Trade and finance cannot really be separated: finance follows trade, and enhanced de facto integration in each feeds the other.
Trade requires currency risk management. When an Indian firm signs a long-term contract to buy/sell with invoicing in Yen, the Indian firm needs to be sure that Japan will stay open so as to enable INR/JPY hedging in the future.
If an MNC makes a direct investment in a country, it needs some assurance that it can bring in funds (equity and loans) to finance the investment, take them out when it wants to run down its operations, and repatriate profits in the meantime. It also needs to be able to hedge its currency exposure.
Hence, entering into trade/investment contracts today is assisted by confidence that liberalisation put in place today will still be there tomorrow.
More generally, there is a big difference between (a) a move today and (b) a move today + a commitment about behaviour tomorrow. Permanent tax cuts yield a much greater consumption response. Permanent capital account liberalisation leads to more FDI and trade. A variety of mechanisms need to be found through which reforms can be given stronger commitment so as to rule out risk of reversal in the future.