Since the start of the financial crisis, central banks across the world have cut interest rates substantially. In Japan, the US, the Eurozone and elsewhere, short-term interest rates controlled by the central bank are at or close to zero.
There are two reasons why this may have happened.
- Macroeconomic conditions worsened drastically after the collapse of Lehman in September 2008.
This prompted central banks, in line with their usual reactions to economic developments, to make correspondingly sharp cuts in interest rates in order to mitigate the collapse in economic activity and to prevent deflation from taking hold.
- More intriguingly, central banks may have perceived a risk that policy rates may eventually reach zero and therefore altered their interest rate setting strategy.
Economic theory suggests that if policymakers grow concerned that the zero lower bound may become binding in the future, it is optimal to cut interest rates below the level they would otherwise have set in order to reduce the likelihood that macroeconomic conditions deteriorate to the point that zero or even negative interest rates become desirable (e.g., Reifschneider and Williams 2000 or Orphanides and Wieland 2000). Or as Giavazzi and Giovannini (2010) put it, they may have been worried about the “low interest-rate trap”.
That said, it has been argued that the possibility of hitting the zero lower bound calls for more caution in cutting rates, so as to retain the possibility to cut interest rates at a later date if the economy worsens further (e.g., Bini-Smaghi 2008). This is the notion of “keeping the powder dry.” A central bank following this strategy would therefore set higher interest rates than it would do in the absence of the zero lower bound.
In short, the issue is whether fear of the zero lower bound producers monetary policy loosening that is more or less aggressive than usual.
New research on the impact of the lower bound on interest-rate setting
The current environment provides a unique opportunity to assess whether and how central banks have responded to the threat of the zero lower bound. In a recent working paper we study whether the ECB, which has let overnight interest rates fall to about 0.25% (that is, much below the repo rate which was cut to 1%), might have cut interest rates more aggressively than it might otherwise have done because of concerns about interest rates in the future reaching zero (Gerlach and Lewis 2010).
The main problem we face is that we do not observe what interest rate the ECB would have set if the zero lower bound was irrelevant. We attempt to overcome this complication by estimating a reaction function for monetary policy on monthly data over the period 1999 to 2009, using the overnight interest rate as the dependent variable. Since the zero lower bound is unlikely to have been a consideration before the financial crisis started in August 2007, but might have been so subsequently, we argue that the reaction function may well have shifted some time in the 2007 to 2009 period.
To account for such a shift, we follow Mankiw et al. (1987) and estimate a smooth transition model that allows for a gradual shift over time between two regimes. We argue that the reaction function in force before the shift provides an estimate of how the ECB responded to macroeconomic conditions in “ordinary” times when the zero lower bound was irrelevant. The reaction function after the shift can then be seen as providing an estimate of the reaction function in “crisis” times when the zero lower bound was a distinct possibility. By comparing the predicted interest rates under the “ordinary” reaction function and the “crisis” reaction function we can then explore whether the ECB’s rate setting behaviour become more aggressive, less aggressive or remained the same across the two periods.
We allow the data to identify the timing and speed of the transition. We find strong evidence of a shift in the autumn of 2008 – just after the collapse of Lehman Brothers – and that the shift took a few months to occur. That is well after the first turbulence in the interbank markets in August 2007, but around the time that most commentators recognised that the failure of Lehman had triggered a large scale financial crisis with obvious implications for the outlook for economic activity and inflation in the Eurozone and elsewhere.
Turning to the two reaction functions, we find that in “ordinary” times the ECB tightens monetary policy in response to stronger economic conditions, higher inflation and a weaker exchange rate. There is also some evidence that the ECB responded to money supply growth, as suggested by its monetary policy framework.
In this “crisis” period, by contrast, the macroeconomic variables are all insignificant and the overnight interest rate follows a first-order autoregressive model. We interpret this result as suggesting that the ECB depressed the overnight interest rate as far as practicable, and let it fluctuate within a narrow range in response to market forces.
Next we ask what level of interest rates we would have expected to observe in the period August 2007 to December 2009 if the “ordinary” reaction function, which captured interest rate setting in periods in which the zero lower bound was not an issue, had remained in force during the full sample. Figure 1 shows our estimate of this interest rate, together with a 95% confidence bound. It indicates that the worsening of macroeconomic conditions would have led the ECB to cut interest rates aggressively after the collapse of Lehmann in September 2008: by around 100 basis points in the last quarter of 2008, and then a further 200 points over the course of 2009. However, the actual cuts in the interest rate came much faster, by around 250 basis points in the last quarter of 2008 alone.
Figure 1. Dynamic forecast of interest rates assuming no regime change
These estimates suggest that the ECB cut interest rates after the collapse of Lehman Brothers considerably more aggressively than we would have expected given an unchanged reaction function. Thus, we interpret them as suggesting that the presence of the zero lower bound did impact on the Governing Council’s policy decisions in line with the theoretical literature on optimal interest rate setting and the zero lower bound.
But while encouraging, the results are also compatible with other interpretations. Orphanides (2010) argues, but provides no estimates, that the ECB’s interest rate setting during the crisis is compatible with a stable reaction function in which forecasts of economic growth and inflation enter. Under that interpretation, the shift in the ECB’s reaction function that we identify above may instead be evidence of a shift in the relationship between current economic conditions and near-term forecasts of inflation and output.
Another possibility is that the true reaction function is non-linear and entails “recession aversion” in the sense of Gerlach (2003). This hypothesis holds that central banks are more concerned – for reasons unrelated to the zero lower bound – by economic activity being below than above the objective. Thus, as economic activity slowed, the ECB started to cut interest rates increasingly aggressively to support growth.
Overall, more work is needed to fully distinguish between the potential role of these different factors in ECB interest rate setting. We view our papers as a first step in that direction.
Bini Smaghi, L (2008), “Careful with the ´d´ words!”, speech given in the European Colloquia Series, Venice, 25 November.
Gerlach, S (2003), “Recession aversion, output and the Kydland-Prescott Barro-Gordon model”, Economic Letters, 81:389-394.
Giavazzi, Francesco, and Alberto Giovannini (2010), “The low-interest-rate trap”, VoxEU.org, 19 July.
Gerlach, S and Lewis, J (2010), “The Zero Lower Bound, ECB Interest Rate Policy and the Financial Crisis”, CEPR Discusison Paper 7933.
Mankiw, N, J Miron and D Weil (1987), “The Adjustment of Expectations to a Change in Regime: A Study of the Founding of the Federal Reserve”, American Economic Review, 77(3):358-374
Reifschneider, D and J Williams (2000), “Three Lessons for Monetary Policy in a Low-Inflation Era”, Journal of Money, Credit and Banking, 32(4):936-966.
Orphanides, A (2010), “Monetary Policy Lessons from the Crisis”, CEPR Discussion Paper 7891.
Orphanides, A and V Wieland (2000), “Efficient Monetary Policy Design Near Price Stability”, Journal of the Japanese and International Economies, 14:327-365.