Are Capital Flows Expansionary or Contractionary? It Depends What Kind

IMF Blog, December 7, 2015

Source details

Canonical URL

Are Capital Flows Expansionary or Contractionary? It Depends What Kind

Other formats

• Markdown version
• Structured JSON version
• Bundle manifest

Bibliographic details

• Authors: Olivier Blanchard, Jonathan D Ostry, Marcos Chamon, Atish Rex Ghosh
• Published: December 7, 2015

Overview

• Authors: Olivier Blanchard, Jonathan D. Ostry, Marcos Chamon, Atish Rex Ghosh
• Date: December 7, 2015
• Central question: Do capital inflows generally expand or contract economic activity in emerging markets?
• Key empirical observation motivating the analysis: In typical emerging-market cases, capital inflows appear to be associated with currency appreciations, credit booms, and output increases (Ostry et al., 2012a).

Theoretical reconciliation: bonds vs. non-bonds

• Standard model implication:
  • Mundell-Fleming model suggests that, for a given monetary policy rate, inflows lead to an appreciation and thus to a contraction in net exports and a decrease in output.
  • Paul Krugman (2013) used a related model to argue that capital outflows are expansionary.
• Extended-asset framework offered by the authors:
  • Distinguishes between “bonds” (policy-rate-type assets) and “non-bonds” (equities, bank liabilities—imperfect substitutes for bonds).
  • Capital inflows can lower the rate on non-bonds even if the policy rate is unchanged, reducing the cost of financial intermediation.
  • The positive domestic-demand effects from lower non-bond rates can offset adverse effects from currency appreciation.
  • Conclusion: Capital inflows may be expansionary even for a given policy rate; this reconciles the Mundell-Fleming intuition with policymakers’ observations.

Role of foreign exchange market intervention

• Sterilized FX intervention through bonds:
  • Can fully offset the effects of bond inflows, leaving both the exchange rate and interest rates unchanged.
  • Mechanism: central bank takes opposite position to foreigners—reduces demand for domestic bonds and increases holdings of foreign assets when foreigners increase demand for domestic bonds.
• Sterilized intervention in response to non-bond inflows:
  • Can avoid currency appreciation, but only at the cost of a larger decrease in the rate of return on non-bonds (see Blanchard et al., 2015).

Capital controls

• Targeted capital controls alter the mix and effects of inflows:
  • Controls on bond inflows:
  • Reduce bond inflows.
  • Increase the effects of non-bond inflows on the exchange rate and on the rate of return on non-bonds.
  • Controls on non-bond inflows:
  • Magnify the effect of bond inflows on the exchange rate.
• In both cases:
  • Targeted capital controls reduce upward pressure on the currency, which increases the “spillover” effects from the non-targeted inflow on the exchange rate, the rate of return, or both.

Effects of monetary policy and the “policy dilemma”

• Central bank objective choices (besides output/inflation mandate):
  • Stabilize the exchange rate → will lower the policy rate.
  • Stabilize the rate of return on non-bonds (limit credit expansion) → will increase the policy rate.
• Policy dilemma framed:
  • Trade-off between stabilizing the exchange rate and limiting declines in non-bond returns.
• False dilemma:
  • A combination of instruments (monetary policy + FX intervention) can, in principle, offset the effects of inflows on both the exchange rate and the rate of return to non-bonds without capital controls (see Ostry et al., 2012b).
  • This offers a more optimistic view than Rey (2013), who argued that, short of macroprudential tools or capital controls, countries cannot divorce themselves from global financial flows.

Empirical evidence

• Identification challenge:
  • Theoretical arguments concern effects of exogenous capital flows; empirical work requires instruments that affect inflows but are plausibly exogenous to domestic events.
  • Empirical analysis must control for countries’ policy tool usage that may offset flow effects.
• Main empirical findings:
  • Bond inflows have a negative effect on economic activity.
  • Non-bond inflows have a significant and positive effect on economic activity.
  • Non-bond inflows (excluding FDI) have a strong positive effect on credit, much stronger than bond flows—highlighting the credit channel as key to output effects.

Policy implications

• Different nature of inflows requires different policy mixes:
  • Bond inflows: more likely contractionary for a given policy rate; sterilized intervention via bonds can neutralize effects on exchange rate and interest rates.
  • Non-bond inflows: can be expansionary by lowering non-bond rates and stimulating credit; sterilized intervention can prevent appreciation but may further depress non-bond returns.
  • Targeted capital controls change the relative pressures from different inflows and can amplify spillovers from the non-targeted inflow type.
  • Combining instruments—monetary policy, FX intervention, and, where appropriate, capital controls or macroprudential measures—allows policymakers to manage exchange-rate and credit-return trade-offs without accepting a strict dilemma.

Source: IMF blog post "Are Capital Flows Expansionary or Contractionary? It Depends What Kind" (December 7, 2015).

Content in this bundle

• 14th Jacques Polak Annual Research Conference; November 7–8, 2013
  • 14th Jacques Polak Annual Research Conference; November 7–8, 2013 (Markdown version)
  • 14th Jacques Polak Annual Research Conference; November 7–8, 2013 (PDF)
• Staff Discussion Note
  • Staff Discussion Note (Markdown version)
  • Staff Discussion Note (PDF)

References

• recent work