Assessing BNM’s Crisis Response

Quite in keeping with yesterday’s announcement, there’s a new working paper from the IMF which looks at the monetary measures implemented by BNM during the 2008-2009 recession and assesses their effectiveness (abstract):

An Assessment of Malaysian Monetary Policy during the Global Financial Crisis of 2008-09
Alp, Harun and Selim Elekdag & Subir Lall

Summary: Malaysia was hit hard by the global financial crisis of 2008-09. Anticipating the downturn that would follow the episode of extreme financial turbulence, Bank Negara Malaysia (BNM) let the exchange rate depreciate as capital flowed out, and preemptively cut the policy rate by 150 basis points. Against this backdrop, this paper tries to quantify how much deeper the recession would have been without the BNM’s monetary policy response. Taking the most intense year of the crisis as our baseline (2008:Q4-2009:Q3), counterfactual simulations indicate that rather the actual outcome of a -2.9 percent contraction, growth would have been -3.4 percent if the BNM had not implemented countercyclical and discretionary interest rate cuts. Furthermore, had a fixed exchange rate regime been in place, simulations indicate that output would have contracted by -5.5 percent over the same four-quarter period. In other words, exchange rate flexibility and the interest rate cuts implemented by the BNM helped substantially soften the impact of the global financial crisis on the Malaysian economy. These counterfactual experiments are based on a structural model estimated using Malaysian data.

I admit to being a little disappointed about the primary results – the 150bp cut in interest rates resulted in a “cushion” worth just 0.5% of GDP. The stunner though is the simulation involving a fixed exchange regime, as the “cushion” from having a floating exchange rate is over four times greater than the cut in policy interest rates.

Or maybe I shouldn’t be surprised, as I’ve noted that policy rate pass through into Malaysian market interest rates is incomplete (here and here). The results of the IMF paper implies that a exchange rate target for monetary policy might be more effective (a’la Singapore), especially bearing in mind Malaysia is hugely exposed to external demand conditions.

Two caveats to bear in mind: peak to trough, the Ringgit fell (in monthly average terms) about 10% for that 2% of GDP “cushion”. So full dependence on an exchange rate target might involve some pretty wild gyrations in the exchange rate, which carries costs of its own (one advantage that an exchange rate target might have though is that you’re not constrained by the zero-interest rate boundary).

A second issue is Goodhart’s law, which states that, “Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes”. Econoenglish translation: if you start using the exchange rate for policy purposes, the effect on the broader economy might differ from what you expect it to be (and at the limit, won’t work at all).

Technical Notes

Alp, Harun and Selim Elekdag & Subir Lall, "An Assessment of Malaysian Monetary Policy during the Global Financial Crisis of 2008-09", IMF Working Paper No. 12/35, January 2012