LOOSE US MONETARY POLICY NOT TO BLAME FOR THE CRISIS

US monetary policy was not too loose during the period 2001-06 and did not contribute to the global financial crisis. That is the conclusion of research by Alessandro Flamini and Costas Milas, presented at the Royal Economic Society’s 2010 annual conference.

Many economists have argued that US interest rates were too low in the run-up to the crisis. For example, compared with the policy rule described by the US economist John Taylor – the Taylor rule – US interest rates between 2001 and 2006 were too low by as much as two percentage points.

But according to the new study, the Taylor rule does not take sufficient account of the uncertainty surrounding the output gap and the inflation gap – and the fact that policy-makers need to move pre-emptively in response to output gap movements. An optimal policy rule that takes account of uncertainty and the need for pre-emptive action fits much more closely than does the Taylor rule with actual interest rates during 2001-06.

More…

The global financial crisis that began in July 2007 has had profound effects on the global economy. The magnitude of the event and the scale of the disruption caused have led to much speculation on the deeper causes of the crisis.

Loose monetary policy in the years before the crisis, especially in the US, has been suggested as one of its major causes. We question the validity of this claim in the context of an optimal policy reaction function in which policy-makers set interest rates by taking into account the level of uncertainty surrounding the business cycle and the true model of the economy.

The view that loose monetary policy in the years before the crisis played a major role in causing the crisis is particularly associated with John Taylor, who has argued that in 2001-06, US interest rates were below those implied by a Taylor rule, by up to 2 percentage points.

Unsurprisingly, Federal Reserve Chairman Ben Bernanke is more sceptical about this claim noting (during the Annual Meeting of the American Economic Society in January 2010) that Taylor does not consider the forward-looking nature of US monetary policy.

This debate has an international dimension because interest rate decisions made by the Federal Reserve Bank set the tone followed by other major central banks including the Bank of England’s Monetary Policy Committee (MPC); the latter discusses in detail the Fed’s decisions in its published MPC minutes.

We add to the debate by comparing the US actual policy interest rate with that implied by a more sophisticated (than previously thought) optimal monetary policy rule. In contrast to previous work, we derive an optimal monetary policy rule not only in terms of the goal variables, that is, inflation gap (i.e. consumer price inflation relative to the 2% target) and the output gap (i.e. real GDP output relative to equilibrium) but also in terms of the uncertainty surrounding these goal variables.

For example, output gap uncertainty corresponds to uncertainty on where the economy is located with respect to the business cycle. This provides an important challenge that policy-makers face when setting interest rates.

Viewing uncertainty as a key factor in assessing whether monetary policy is optimal, we show that when the uncertainty surrounding the output gap increases, the optimal policy response to that variable increases too. Intuitively, when output is currently some distance from its equilibrium level, increased output uncertainty is likely to widen the gap between output and equilibrium even further in the immediate future.

To prevent this from happening, policy-makers move in a pre-emptive manner by increasing their response to output gap movements.

The predictions of our optimal policy rule are tested on real-time US data. Figure 1 plots the US policy interest rate together with the policy rate derived from our optimal policy rule under uncertainty. Since 2000, the two interest rate series have moved very close to each other.

In fact, the actual US policy interest rate has been, on average, only 25 basis points lower than the optimal policy rate over the 2001-06 period; this difference is not statistically significant. The implication of our analysis is that once uncertainty about the goal variables is properly taken into accounted, US monetary policy turns out to be optimal during the pre-crisis period.

To limit the impact of a (potential) future financial crisis, it is necessary to assess what went wrong but also do justice on policies that did not add to the 2007-09 crisis. Our work rejects the claim that US monetary policy has been loose enough to contribute to the global financial crisis.

ENDS

Note for editors: ‘Real-time Optimal Monetary Policy with Undistinguishable Model Parameters and Shock Processes Uncertainty’ by Alessandro Flamini (University of Sheffield) and Costas Milas (Keele University) was presented at the 2010 RES conference, Surrey University.

Contact: Alessandro Flamini 079 400 070 or 0039 389 031 8462 (email: ); Costas Milas 078 878 87069 (email:)

Figure 1: US optimal policy interest rate and actual policy interest rate, 2000-09