This year marks the 20th anniversary of the Asian Financial Crisis. Many factors were involved in that disaster, but grossly excessive foreign capital inflows were the key macro-economic problem during the boom years preceding the crisis. These flooded out when 'euphoria turned to panic without missing a beat'. Much analysis has gone into finding the right policy response, but the answer remains elusive.

This policy lacuna is likely to be tested further over coming months, when US President Donald Trump's tax reforms may bring US capital back onshore and the US Federal Reserve begins to unwind its bloated balance sheet resulting from quantitative easing.

The textbook answer is to absorb the inflows by letting the exchange rate rise. Policy-makers in emerging economies, however, find this an unappealing response. Their relatively tiny financial markets can be overwhelmed by the inflows from much larger volatile portfolio adjustments in advanced countries. The exchange rate appreciation required to equilibrate the inflow would often be very substantial, undermining international competitiveness and shrinking the export sector. Why would it make sense to contract what is often the most dynamic part of the economy, to make room for foreigners to flood into asset markets (shares and property), setting off a disruptive asset-price boom, only to have these inflows reverse when foreign sentiment changed?

One response is to offset the inflow by intervening heavily in the foreign exchange market, restraining the exchange rate appreciation and building up reserves to handle the later outflow. But this means that the central bank is effectively buying up the inflow and reinvesting it at a lower interest rate in foreign reserves, ready to facilitate the foreigners' later flight. That can't be a profitable deal for the host country.

A recent IMF paper captures the state of policy disarray. The high-profile authors see a 'natural mapping' that suggests tailor-made policy responses, depending on the nature of the inflow. The old recommendation (just let the exchange rate appreciate) is no longer put forward as the panacea, though it is still a firm favourite among many of their colleagues at the IMF. Perhaps endorsing the old jibe that 'IMF' stands for 'It's Mainly Fiscal', these authors see fiscal tightening as the most-likely logical answer. More in sorrow than in anger they observe, however, that in their empirical study of 30 emerging economies, hardly anyone does this:

The orthodox policy prescription to tighten fiscal policy in the face of capital inflows was the least used instrument in practice, with no strong evidence that EMEs systematically tightened fiscal policy in response to large capital flows.

Why don't policy-makers in the emerging economies share this preference for fiscal tightening? Tighter fiscal policy might be appropriate if the domestic economy was already overheating. But as a response to excessive capital inflow, it doesn't make much sense. Why so? Tighter fiscal policy increases overall domestic saving – this reduces the current account deficit (the current account is equal to the savings-investment balance, by national accounts identity). The capital inflow has to equal the now-smaller current account deficit, and this can be brought into equilibrium only if the floating exchange rate appreciates, discouraging both capital inflow and exports. Thus we are back with the same old problem: the dynamic export sector has to make room to accommodate the foreigners. As well, the fiscal tightening will have a cost in terms of foregone expenditure or higher taxes. All this just to allow foreigner investors to come in, pushing up asset prices and setting off a credit boom!

Perhaps practical policy-makers might be tempted to cut the Gordian knot by imposing capital controls, especially for short-term volatile capital inflows (leaving characteristically-stable foreign direct investment untouched). The IMF has come some distance from its earlier free-market dogma on these issues. Before 1997, capital inflows were seen as unambiguously beneficial to all. Capital controls, whether on inflows or outflows, were anathema. Now they are included in the policy tool-box, but only on a last-resort basis.

There is still a long way to go, and any policy-maker tempted to use 'capital flow management' (the acceptable face of 'capital controls') would get lukewarm encouragement from the IMF. For some of us, the puzzle raised by this study of 30 economies is not why they failed to use fiscal policy, but why they remain reluctant to use capital controls vigorously, starting with substantial taxes on short-term footloose inflows.