Blanchard and others look to reconcile theory of capital flows with practice

A group of economists has found what may be a solution to the "striking" divide between the theoretical and actual impact of capital flows on an economy.

Former International Monetary Fund chief economist Olivier Blanchard and IMF economists Marcos Chamon, Atish Ghosh and Jonathan Ostry express surprise this is still an issue for economics. "One would think that the question was settled long ago," they say. "But, in fact, it is not."

They approach the problem using an updated version of the Mundell-Fleming model, a standard textbook model of small, open economies.

The problem with the traditional Mundell-Fleming model is it predicts capital inflows to be contractionary, because they appreciate the currency and reduce net exports. This sharply contrasts with the experience of many emerging markets, which tend to associate capital inflows with booms, the authors argue in a working paper published by the US-based National Bureau of Economic Research.

By dividing assets in the economy into 'bonds' and 'non-bonds', the latter group encompassing stocks, foreign direct investment and bank deposits, the authors are able to differentiate between types of flows and types of policy responses, to give a more nuanced account.

They assume the central bank's policy rate to be equal to the rate on bonds, which means even holding that rate constant, inflows can decrease rates on non-bonds, lowering the costs of financial intermediation and offsetting the effects of exchange rate appreciation.

Thus, although bond inflows are contractionary, non-bond inflows may be expansionary. This in turn means that policy responses, namely sterilised foreign exchange interventions or capital controls, will have different effects depending on which kind of flow they target.

The authors note calculating the correct combination of policies is beyond the scope of the paper, as it does not consider distortions such as credit growth. Encouraging too much non-bond growth could create a credit boom and subsequent economic slump.

However, they tentatively conclude, "if the country wants to increase output, and has macro-prudential tools to avoid excessive credit growth, then non-bond flows are more attractive than bond flows".

Empirical evidence

The authors look for evidence of the effects of capital flows in 19 emerging market economies using data from 2000 onwards, finding the model's results appear to fit with the experience of emerging markets.

Pinning down the exogenous effects proves tricky, and the authors use instruments first to isolate flows based on global rather than domestic factors, and second to examine only the direct effects of flows, not second-round effects that follow capital flow policies. They also use dummy variables to control for countries that are deemed safe havens.

The results imply the effect of bond flows is negative and insignificant, while the effect of non-bond flows is positive and significant, "both statistically and economically". An increase in exogenous non-bond flows of 1% of GDP increases GDP growth by 0.31 percentage points.

Furthermore, disaggregating non-bond flows into foreign direct investment, portfolio equity flows and "other" flows shows all three display positive, significant results of roughly the same magnitude.

The authors describe their empirical results as "very much a first pass", and call for further research. "But, overall, we see the set of results as strongly supportive of the distinction between bond and non-bond flows," they add.